DOT consists of multiple modal Operating

UNITED STATES DEPARTMENT OF
TRANSPORTATION
GLOBAL POSITIONING SYSTEM (GPS)
ADJACENT BAND COMPATIBILITY ASSESSMENT
I
FINAL REPORT
APRIL 2018

II
EXECUTIVE SUMMARY
The goal of the U.S. Department of Transportation (DOT) Global Positioning System (GPS)
Adjacent Band Compatibility Assessment is to evaluate the maximum transmitted power levels
of adjacent band radiofrequency (RF) systems that can be tolerated by GPS and Global
Navigation Satellite System (GNSS) receivers. The results of this effort advance the
Department’s understanding of the extent to which such adjacent band transmitters impact
GPS/GNSS devices used for transportation safety purposes, among numerous other civil
applications. The assessment described in this report addresses transmitters in bands adjacent to
the 1559-1610 MHz radionavigation satellite service (RNSS) band used for GPS Link 1 (L1)
signals that are centered at 1575.42 MHz.
The assessment includes two primary components:
• One component, led by the DOT Office of the Assistant Secretary for Research and
Technology (OST-R), focused on all civilian GPS devices and their applications, apart
from certified aviation. Through this component of the Study, categories of receivers were
evaluated that included aviation (non-certified), cellular, general location/navigation, high
precision, timing, and space-based receivers. An element of this effort was to determine
equipment susceptibility to adjacent band interference to support analyses for deriving
compatible power levels.
• The other component, led by the Federal Aviation Administration (FAA), focused on
certified GPS avionics, and was conducted by analysis to determine the adjacent band
power levels that conform to existing certified GPS aviation equipment standards.
The DOT GPS Adjacent Band Study is the product of an extensive process to gather stakeholder
views and input. OST-R and FAA benefited significantly from feedback received via
governmental and public outreach on equipment use cases, interaction scenarios, propagation
models, and transmitter characteristics.
Certified GPS avionics meet their performance requirements when operating within the RF
interference (RFI) environment defined in appropriate FAA Technical Standard Orders (TSOs).
For civil GPS/GNSS receivers other than certified avionics, receiver testing needed to be
conducted to determine the Interference Tolerance Masks (ITMs) for various categories of
receivers. ITM defines, for a particular receiver, the maximum received aggregate interference
power that can be tolerated by the corresponding tested GPS/GNSS receiver.
To accomplish this testing, OST-R sought to include a broad range of devices used in rail,
aviation, motor vehicle, maritime, and space applications, among a number of other civil uses of
GPS/GNSS including timing, surveying, precision agriculture, weather forecasting, earthquake
monitoring, and emergency response. The GPS/GNSS receivers for this test effort were provided
by U.S. Government and industry partners and represented the diverse nature of GPS/GNSS
applications and services.

III
GPS/GNSS receiver testing, led by the OST-R/Volpe Center, was conducted at the U.S. Army
Research Laboratory (ARL) at the White Sands Missile Range (WSMR) facility in New Mexico
in April of 2016 with 80 civil GPS and GNSS receivers tested, as shown in Figure ES-1. The Air
Force GPS Directorate conducted testing of military GPS receivers the week prior to the civil
receivers being tested.
Figure ES-1: GPS/GNSS Receivers in WSMR Anechoic Chamber
In determining the transmit power level analysis, it is important to understand real-world
scenarios and the proximity those applications of GPS/GNSS may come to adjacent band
transmitters. A graphic of various emergency response uses is shown in Figure ES-2. First
responders are increasingly using GPS/GNSS to locate patients both during emergencies and as a
normal course of duty. As shown in the figure, there are multiple uses of GPS/GNSS for
navigation of emergency service response vehicles, as well as asset tracking, including increased
situational awareness of where response personnel and vehicles are located. An unmanned
aircraft system (UAS) or drone, which also has a GPS/GNSS receiver incorporated also plays a
role in this scenario, supporting the response effort. Drones are becoming of increasing
importance in collecting imagery and sensor data in response to natural disasters and other
incidents.
This scenario illustrates that use of a GPS/GNSS receiver can be quite close in distance -- within
tens of meters of a base station transmitter and potentially very close to a handset as well
transmitting in the adjacent band. The GPS/GNSS receiver also could be located vertically
above the base station.

IV
Figure ES-2: Emergency Response Use Case
Results for the high precision receiver category for an emitter at 1530 MHz based on results of
analysis and testing are presented in Figure ES-3. These results are for a typical cellular base
station power level of 29 dBW (794 watts) with the base station antenna 25 m above the ground.
In this figure, the horizontal axis is the lateral distance between the GPS/GNSS receiver and the
base station. The vertical axis is the height of the GPS/GNSS receiver above the ground. Note
the high precision category of receiver exceeds a 1 dB signal-to-noise density (C/N0) interference
protection criteria at a distance beyond 14 km from the transmitter. When this occurs, the
behavior of the GPS/GNSS receiver can become unpredictable in its ability to meet the accuracy,
availability, and integrity requirements of its intended application and a receiver in a mobile
application may not be able to reacquire GPS positioning as the mobile application encounters
multiple, closely-spaced emitters in an urban scenario. Furthermore, this category of receiver
experiences loss of lock for low elevation GPS/GNSS satellites at distances up to 3 km with loss
of lock on all satellites at approximately 1 km from the transmitter.
Figure ES-3: Impact of a 29 dBW Cellular Base Station Transmitting at
1530 MHz on a High Precision GPS/GNSS Receiver
0 2000 4000 6000 8000 10000 12000 14000
Distance from Tower (m)
0
1000
2000
Height
(m)
≥ 1 dB C/N
0
degradation
Loss of Lock on Low Elevation Satellites
Loss of Lock on All Satellites

V
Further analysis was performed to determine the maximum tolerable power levels for various
categories of civil GPS/GNSS receivers for deployments of a macro urban and micro urban
cellular network at frequencies within 100 MHz of GPS L1 (1475 – 1675 MHz). As an example,
the results for 1530 MHz are shown in Table ES-1 for general location and navigation (GLN),
high precision (HPR), Timing (TIM), and cellular (CEL) receivers. The transmit power level as
quantified by the effective isotropic radiated power (EIRP) that can be tolerated is a function of
distance from the transmitter. Two distances were chosen for evaluation (10 m and 100 m). The
results demonstrate that other than the cellular devices, the other categories of GPS/GNSS
receivers are sensitive to adjacent band power and can tolerate levels in the milliwatts or
microwatts range as described below, depending on the separation distance to the transmitter.
Table ES-1: Maximum Tolerable Power Level
for GPS/GNSS Receivers at 1530 MHz
Table ES-2 depicts the maximum tolerable power levels of space-based receivers used for
performing scientific measurements. A future NASA mission, COSMIC-2, fitted with a TriG
receiver built by NASA/Jet Propulsion Laboratory, was modeled, simulated, and analyzed using
various cellular network deployment scenarios. The COSMIC-2 mission will be operating at an
orbit of 800 km.

VI
Table ES-2: Maximum Tolerable Power Level
for Space-Based Receivers at 1530 MHz
For certified GPS avionics, the FAA analyzed a number of scenarios including:
1) Inflight Aircraft with a Ground-based Handset
2) Inflight Aircraft with a Ground Base Station
3) Inflight Aircraft with an Onboard Handset
4) Aircraft on the ground with an Onboard Handset
5) Aircraft at Gate / Single Handset Source on or near Boarding Stairs or Jetway
6) Aircraft at Gate/Users Inside Airport
7) Terrain Awareness Warning System (TAWS) / Helicopter TAWS (HTAWS) Scenarios
with Ground-based Mobile Broadband Handsets
8) TAWS and HTAWS Scenarios with Broadband Base Station
The analysis for certified avionics is based on the concept of an “assessment zone” (see Figure
ES-4) inside of which GPS performance may be compromised or unavailable and GPS-based
safety systems will be impacted accordingly due to the elevated levels of RFI. Under the
described engineering and operational assumptions, helicopter operations are the limiting factor
in the analysis. These analyses indicate that protection of certified avionics, operating under the
assumption of the described 250 foot (76.2 m) radius assessment zone, requires that the ground
station transmission not exceed 9.8 dBW (10W) (cross-polarized) at 1531 MHz. This limit is
obtained from the HTAWS scenario which was found to be the most restrictive of the certified
aviation scenarios examined.

VII
Figure ES-4: Candidate Assessment Zone (Not to Scale)
This concept generated a number of comments and questions from the aviation community when
vetted through RTCA, Inc. One rotorcraft operator stated that its pilots use visual reference
within the assessment zone and the assessment zone would have no negative impact on their
operation. However, there were unresolved concerns expressed by several, though not all,
operators about the assessment zone and its impacts to aviation operations and safety.
These concerns include: technical and human factors issues associated with re-initialization of
GPS after loss of the signal or when the signal reception is intermittent; workload and human
factors impacts on pilots to monitor and track assessment zone locations; the possibility that pilot
workload, confusion, or error could lead to aircraft inadvertently entering an assessment zone
and losing needed GPS functionality; and impacts to onboard and ground systems that are
dependent upon GPS, such as Automatic Dependent Surveillance (ADS) Broadcast/Contract
(B/C), or fixed-wing and helicopter terrain awareness warning system including obstacle
alerting.
The FAA has not completed an exhaustive evaluation of the operational scenarios in developing
this assessment zone. Further, the current analyses do not include an operational assessment of
the impact of the assessment zone in densely populated areas, which may present additional
variables, including the risk posed to people and property for operations such as UAS using
certified avionics which may be required to operate within the assessment zone.

VIII
However, based on the results of the OST-R testing and analysis of the other categories of
receivers, the transmitter power level that can be tolerated by certified aviation may cause
interference with, or degradation to, most other categories of GPS/GNSS receivers including
those used for General Aviation and drones, as detailed in the results set forth in this report.
The U.S. Department of Transportation would like to thank all of the Federal departments and
agencies for their participation in this effort, including the National Telecommunications and
Information Administration (NTIA) and Federal Communications Commission (FCC), as well
the GPS/GNSS receiver manufacturers who participated in the testing, and all of the stakeholders
who attended the public workshops and RTCA meetings and provided valuable feedback during
this effort.

IX
TABLE OF CONTENTS
EXECUTIVE SUMMARY ............................................................................................................II
TABLE OF CONTENTS..............................................................................................................IX
LIST OF FIGURES ................................................................................................................... XIII
LIST OF TABLES..................................................................................................................... XVI
1. INTRODUCTION................................................................................................................. 17
2. BACKGROUND................................................................................................................... 20
3. Civil Receiver Testing........................................................................................................... 22
3.1 Anechoic/Radiated Testing ............................................................................................ 22
3.1.1 Devices Under Test (DUTs)....................................................................................... 22
3.1.1.1 GPS/GNSS Receivers Tested.............................................................................. 23
3.1.1.2 Antennas Tested.................................................................................................. 25
3.1.2 Anechoic Chamber..................................................................................................... 26
3.1.3 Location in Chamber.................................................................................................. 26
3.1.3.1 GNSS Signal Generation..................................................................................... 28
3.1.3.2 Interference Signal Generation ........................................................................... 32
3.1.3.2.1 Type 1 Signals .............................................................................................. 34
3.1.3.2.2 Type 2 Signals .............................................................................................. 35
3.1.3.2.1 Intermodulation ............................................................................................ 36
3.1.4 System Calibration and Chamber Mapping ............................................................... 37
3.1.4.1 GNSS System...................................................................................................... 37
3.1.4.2 SPIGAT............................................................................................................... 40
3.1.5 Test Sequence............................................................................................................. 40
3.1.5.1 Linearity Test ...................................................................................................... 41
3.1.5.2 Interference Test.................................................................................................. 41
3.1.6 Data Processing/ITM formation................................................................................. 42
3.1.6.1 Data Conversion and Format .............................................................................. 43
3.1.6.2 1 dB CNR Degradation ....................................................................................... 44
3.1.6.3 ITM Data Processing........................................................................................... 45
3.1.7 ITM Aggregation and Test Results ............................................................................ 50
3.1.8 Loss of Lock Data Processing.................................................................................... 55

X
3.2 Conducted (Wired) Testing............................................................................................ 58
3.2.1 Devices Test ............................................................................................................... 58
3.2.2 Signal Generation....................................................................................................... 59
3.2.2.1 Signal Acquisition............................................................................................... 60
3.2.2.2 Out-of-Band Emissions....................................................................................... 60
3.2.3 System Calibration ..................................................................................................... 62
3.2.4 Test Sequence............................................................................................................. 63
3.2.5 Data Processing .......................................................................................................... 63
3.2.5.1 Comparison Tests................................................................................................ 63
3.2.5.2 OOBE Results ..................................................................................................... 64
3.2.5.3 Acquisition Results ............................................................................................. 65
3.3 Antenna Characterization............................................................................................... 67
3.3.1 Selected Antennas ...................................................................................................... 68
3.3.2 Chamber Measurements ............................................................................................. 68
3.3.3 Live-Sky Measurements............................................................................................. 71
3.3.4 Bench Test Measurements (Active Sub-assembly Measurements)............................ 73
4. Transmit Power Level Analysis (Excluding Certified Aviation) .......................................... 75
4.1.1 Approach .................................................................................................................... 75
4.1.2 Network Transmitter Parameters................................................................................ 75
4.1.2.1 Base Stations ....................................................................................................... 75
4.1.2.2 Handsets .............................................................................................................. 78
4.1.3 Use Case Development............................................................................................... 78
4.1.3.1 Receiver Antenna Patterns .................................................................................. 81
4.1.4 Propagation Models.................................................................................................... 85
4.1.5 Forward Modeling Results and Sensitivity ................................................................ 89
4.1.6 Inverse Transmit Power Calculation Results and Sensitivity Analysis...................... 91
4.1.6.1 Inverse Transmit Power Calculation Results ...................................................... 91
4.1.6.2 Sensitivity Analysis............................................................................................. 96
4.1.6.2.1 Aggregation Effects...................................................................................... 96
4.1.6.2.2 Effects of Propagation Models ..................................................................... 99
4.1.6.2.3 EIRP masks for Median ITMs.................................................................... 102
4.1.7 Summary of Transmit Power Level Calculation...................................................... 103

XI
4.2 Spaceborne and Science Applications.......................................................................... 106
4.2.1 Radio Occultation (GNSS-RO)................................................................................ 107
4.2.2 NASA/JPL TriG Receiver Overview....................................................................... 108
4.2.3 Spaceborne Receiver Assessment for Science-Based Applications......................... 108
4.2.3.1 Summary of TriG Receiver System Characteristics Used for Analysis............ 109
4.2.3.2 Terrestrial LTE Deployment Scenarios ............................................................ 110
4.2.3.3 Summary of BS Transmitter System Characteristics Used for Analysis.......... 113
4.2.3.4 TriG Receiver Analysis..................................................................................... 113
4.2.3.5 Results............................................................................................................... 115
5. Certified Aviation RecEIver................................................................................................ 118
5.1 Determination of Tolerable Interference Levels .......................................................... 118
5.1.1 Area of Aviation Operation...................................................................................... 119
5.1.2 Tracking and Acquisition Thresholds ...................................................................... 121
5.1.2.1 Receiver Tracking Limit Criteria for Adjacent-Band RFI................................ 121
5.1.2.2 Receiver Acquisition Limit Criteria for Adjacent-Band RFI............................ 122
5.1.2.3 Receiver Tracking Limit Criteria for Broadband Handset RFI In-band to GPS123
5.1.3 Transmitter and Receiver Component Assumptions................................................ 123
5.1.3.1 GPS Receive Antenna Gain .............................................................................. 123
5.1.3.2 Broadband Wireless Base Station and Mobile Handset Characteristics........... 125
5.1.3.2.1 Broadband Wireless Base Station Characteristics...................................... 125
5.1.3.2.2 Broadband Wireless Mobile Handset Characteristics ................................ 126
5.2 Transmit Power Level Calculations ............................................................................. 126
5.2.1 Use Case/Interaction Scenario Development ........................................................... 127
5.2.1.1 Inflight Aircraft/Ground-Based Source Scenario Set ....................................... 127
5.2.1.1.1 Inflight Aircraft/Ground-Based Handset Cases.......................................... 127
5.2.1.1.2 Inflight Aircraft/Ground-Based Base Station Cases................................... 128
5.2.1.1.3 Inflight Aircraft / Discretely-located Ground Base Station Cases ............. 128
5.2.1.1.4 Inflight Aircraft / Randomly-located Ground Base Station Cases ............. 128
5.2.1.2 TAWS/HTAWS and Low Altitude Positioning and Navigation Scenarios...... 128
5.2.1.2.1 TAWS / HTAWS and Pos/Nav Scenarios with Ground-based Mobile
Broadband Handsets..................................................................................................... 129
5.2.1.2.2 TAWS and HTAWS Scenarios with Broadband Base Stations................. 129
5.2.1.3 Handset Sources on Board Aircraft................................................................... 129

XII
5.2.1.3.1 Onboard Handset Operation for Aircraft Inflight....................................... 129
5.2.1.3.2 Onboard Handset Operation for Aircraft on Ground.................................. 130
5.2.1.4 Aircraft at Gate Scenarios ................................................................................. 130
5.2.1.4.1 Aircraft at Gate / Single Handset Source on or near Boarding Stairs or
Jetway…………........................................................................................................... 130
5.2.1.4.2 Aircraft at Gate/30 Users Inside Airport .................................................... 130
5.2.2 Propagation Models.................................................................................................. 131
5.2.2.1 Single Path Propagation Model......................................................................... 131
5.2.2.2 Aggregate Effects Model .................................................................................. 132
5.2.3 Tolerable Transmit Power Calculation Results and Sensitivity Analysis................ 133
5.2.3.1 Tolerable Transmit Power Calculation Method Overview ............................... 133
5.2.3.1.1 Tolerable Transmit Power Calculation Method – Base Station Cases....... 133
5.2.3.1.2 Tolerable Transmit Power Calculation Method – Handset Cases.............. 134
Results for Inflight Aircraft/Ground-based Handset Cases............................... 134
Results for Inflight Aircraft / Ground Based Base Station Cases ..................... 135
Results for Onboard Handset Operation for Aircraft on Ground...................... 139
Results for Aircraft at Gate/30 Users Inside Airport ........................................ 140
Results for Inflight Aircraft TAWS/HTAWS and Low Altitude Pos/Nav ....... 142
Frequency Dependencies .................................................................................. 147
Sensitivity Analysis Results.............................................................................. 148
Certified Aviation Receiver Analysis Results Summary.................................. 152
6. SUMMARY......................................................................................................................... 155
ACRONYM LIST....................................................................................................................... 159
REFERENCES ........................................................................................................................... 163
ACKNOWLEDGEMENTS........................................................................................................ 164

XIII
LIST OF FIGURES
Figure 3-1: Chamber Dimensions and Layout (Top View).......................................................... 26
Figure 3-2: DUT and Calibration Grid Locations......................................................................... 27
Figure 3-3: Photo of the DUTs on the test grid ............................................................................ 28
Figure 3-4: GNSS Signal Generation and Recording................................................................... 29
Figure 3-5: GNSS Signal Playback and Transmission ................................................................. 30
Figure 3-6: Interference Frequencies and Signal Levels Tested................................................... 33
Figure 3-7: Interference System Configuration for Radiated Test ............................................... 33
Figure 3-8: Measured gain response: (a) bandpass RF filter with tighter rejection requirements,
(b) bandpass RF filter with more relaxed rejection requirements................................................. 34
Figure 3-9: Type 1 Signal Captured during WSMR Testing @ 1530 MHz................................. 35
Figure 3-10: Type 2 Signal Captured during WSMR Testing @ 1530 MHz............................... 36
Figure 3-11: Intermodulation Signal Captured during WSMR Testing ....................................... 37
Figure 3-12: Measured Gain of Two MiniCircuits ZRL-2400-LN Amplifiers............................ 38
Figure 3-13: GNSS Signal Received Power Variation across the Test Grid (1227 MHz)........... 39
Figure 3-14: GNSS Signal Received Power Variation across the Test Grid (1561 MHz)........... 39
Figure 3-15: Power Correction Representations for Three Frequencies ..................................... 40
Figure 3-16: ITM Processing Block Diagram .............................................................................. 45
Figure 3-17: Sample plot for calibrated interference power overlaid with time aligned CNR data
for a given DUT at a particular interference frequency................................................................ 46
Figure 3-18: Determining the tolerable interference level from the CNR versus interference
power for a one PRN after time alignment and calibration of interference power....................... 47
Figure 3-19: Overlaid L1 C/A ITMs from two radiated LTE test events for a single DUT. Test-2
and Test-3 refer to the first and second LTE tests respectively.................................................... 47
Figure 3-20: CDF of measurement uncertainty calculate from per DUT differences across PRNs
(black) and test to test difference (red) for the 10 MHz LTE interference signal ........................ 49
Figure 3-21: CDF of measurement uncertainty calculate from per DUT differences across PRNs
(black) and test to test difference (red) for the 1 MHz AWGN interference signal ..................... 49
Figure 3-22: GPS L1 C/A bounding ITM for each category of receivers .................................... 50
Figure 3-23: HPR bounding ITMs for each of the emulated GNSS signals................................. 51
Figure 3-24: HPR Bounding ITMs for each of the emulated GNSS signals. ITM bounding masks
for the 1 MHz AWGN and 10 MHz LTE interference signals are shown ................................... 52
Figure 3-25: 10 MHz Statistical Mask Results for High Precision receivers: (a) GPS L1 C/A (b)
All Emulated GNSS Signals......................................................................................................... 53
Figure 3-26: Determination of Loss of Lock Interference Level from CNR Data....................... 54
Figure 3-27: Interference Power resulting in Loss of Lock for GPS L1 C/A-code (High Elevation
Angle). .......................................................................................................................................... 56
Figure 3-28: Interference Power resulting in Loss of Lock.......................................................... 57
Figure 3-29: Interference System Configuration for Wired Tests................................................ 59
Figure 3-30: OOBE Levels Associated with LTE Signal Power used in Testing ........................ 60

XIV
Figure 3-31: Comparison of IP causing 1 dB degradation for the LTE Interference Signal from
Radiated and Wired Testing. (a) Shows a High Precision receiver and (b) shows a Cellular
device............................................................................................................................................ 62
Figure 3-32: Interference power causing 1 dB CNR degradation for baseline and OOBE tests.
(a) Shows a High Precision receiver and (b) shows a Cellular device. ........................................ 63
Figure 3-33: Summary acquisition performance for 1525 MHz. (a) Number of DUTs (b)
average acquisition time for ICD minimum and low elevation satellites..................................... 64
Figure 3-37: Frequency Selectivity of the 14 External Antennas................................................. 68
Figure 3-38: Relative RHCP Gain Patterns of the 14 Antennas at 1575 MHz (red vertical lines
correspond to 5 deg elevation angle) ............................................................................................ 68
Figure 3-39: Relative L1 RHCP Antenna Gain Estimated from Live-sky C/N0 Measurements for
Three GLN Integrated Antennas and Quadratic Fit...................................................................... 69
Figure 3-40: Relative L1 RHCP Antenna Gain Estimated from Live-sky C/N0 Measurements for
an Integrated HPR Antenna and Quadratic Fit ............................................................................. 70
Figure 3-41: Live-sky C/N0 Measurements for a CEL Device..................................................... 70
Figure 3-42: Filter/LNA Responses measured with Bench Testing............................................. 71
Figure 4-1: Macrocell Radius and Intersite Distance are A and B, respectively. Each hexagon is
referred to as either a sector or cell............................................................................................... 74
Figure 4-2: Macro Base Station Antenna Gain Patterns (top – elevation; bottom – azimuth)..... 75
Figure 4-3: Small cell Base Station Antenna Gain Patterns (elevation patterns shown; both
patterns are omnidirectional in azimuth) ...................................................................................... 75
Figure 4-4: Emergency Response Use Case................................................................................. 78
Figure 4-5: Relative VPOL Antenna Gain Patterns for 1530 MHz.............................................. 81
Figure 4-6: Gain Patterns Illustrating Generation of GLN Coefficients for 1530 MHz............... 82
Figure 4-7: Illustration of use case analysis region ...................................................................... 86
Figure 4-8: Macro Urban Base Station (EIRP = 59 dBm), Bounding GAV, 1540 MHz............. 87
Figure 4-9: Maximum Impacted Lateral Distance for Bounding GAV, Macro Urban Base Station
(EIRP = 59 dBm) .......................................................................................................................... 88
Figure 4-10: Macro Urban Base Station (EIRP = 59 dBm), Bounding HPR, 1530 MHz............ 88
Figure 4-11: (a) Tolerable 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑟𝑟, 𝑓𝑓) map in the vertical computation domain, (b)
Tolerable 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑋𝑋, 𝑓𝑓) as a function of standoff distance X......................................................... 90
Figure 4-12: Tolerable 𝐸𝐸𝐼𝐼𝐼𝐼𝐼𝐼(𝑋𝑋, 𝑓𝑓) as a function of standoff distance X up to X=500 m .......... 91
Figure 4-13: EIRP(f,ds=10m) for the HPR category: L1 C/A, micro urban deployment, bounding
EIRP Mask, and FSPL propagation.............................................................................................. 92
Figure 4-14: EIRP(f,ds=10m) for five receiver categories of receivers: L1 C/A, micro urban
deployment, bounding EIRP Mask, and FSPL propagation......................................................... 93

XV
Figure 4-15: Comparison of EIRP(f,ds=10m) L1 C/A and All GNSS masks for the HPR category
of receivers: Micro urban deployment, bounding EIRP Mask, and FSPL propagation ............... 93
Figure 4-16: EIRP(f,ds=10m) for five categories of receivers: All GNSS, micro urban
deployment, bounding EIRP Mask, and FSPL propagation......................................................... 94
Figure 4-17: micro deployment used for the aggregation sensitivity analysis. A small cell of
radius 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟, and transmitters’ interspacing distance ISD. ........................................................... 95
Figure 4-18: (a) Overlay of EIRP(X,f) as a function of standoff distance X for the case of single
and multiple base stations (b) Difference between EIRP(X,f) for the two cases ...................... 96
Figure 4-19: Overlay of 𝐸𝐸𝐼𝐼𝐼𝐼𝐼𝐼(𝑋𝑋, 𝑓𝑓) tolerance masks for the case of a single base station and that
of multiple transmitter case........................................................................................................... 97
Figure 4-20: Tolerable EIRP levels for the case of two ray path loss propagation model (a)
Tolerable 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑟𝑟, 𝑓𝑓) map in the vertical computation domain, (b) Tolerable 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑋𝑋, 𝑓𝑓) as a
function of standoff distance X..................................................................................................... 98
Figure 4-21: (a) Overlay of 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑋𝑋, 𝑓𝑓) as a function of standoff distance X for the case of FSPL
and two ray path loss propagation (b) Difference in tolerable EIRP(X,f) due to propagation
models........................................................................................................................................... 99
Figure 4-22: Comparison between two ray and FSPL EIRP tolerance masks 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑋𝑋, 𝑓𝑓) for
X=100m standoff distance.......................................................................................................... 100
Figure 4-23: EIRP levels corresponding to L1 C/A median ITMs............................................. 100
Figure 4-24: Time Difference of Arrival of GNSS Signal ......................................................... 106
Figure 4-25: Example Satellite View of the U.S. Cities............................................................. 107
Figure 4-26: Earth Station Deployment Zone Model (Report ITU-R SA. 2325-0)................... 110
Figure 5-1: Candidate Assessment Zone .................................................................................... 119
Figure 5-2: CW Interference Susceptibility vs. Frequency, Tracking Mode.............................. 120
Figure 5-3: Lower Hemisphere Installed V-pol and H-pol Receive Antenna Patterns Max. Gain
vs. Elevation Angle..................................................................................................................... 122
Figure 5-4: Upper Hemisphere Installed V-pol. And H-pol. Receive Antenna Patterns Max. Gain
vs. Elevation Angle..................................................................................................................... 123
Figure 5-5: Handset Scenario Probabilities ................................................................................ 133
Figure 5-6: WIRSO Banking Scenario 1 – P(z) values Using Two Methods ............................ 136
Figure 5-7: Aggregate Handset Signal Loss ............................................................................... 138
Figure 5-8: Aircraft at Gate with Thirty Uniformly Distributed Handsets in Terminal ............. 138
Figure 5-9: (1-CDF) Aggregate Power Factor............................................................................ 139
Figure 5-10: HTAWS Dual Polarization 20 m Emitter Antenna-Mean Limits.......................... 146
Figure 5-11: (1-CDF) for Most Restrictive Mean Limit Condition of Figure 5-10.................... 147
Figure 5-12: HTAWS Dual Polarization 10 m Emitter Antenna-Mean Limits.......................... 147
Figure 5-13: (1-CDF) for Most Restrictive Mean Limit Condition of Figure 5-12.................... 148
Figure 5-14: HTAWS Vertical Polarization 25 m Emitter Antenna-Mean Limits..................... 148
Figure 5-15: (1-CDF) for-Most Restrictive Mean Limit Condition of Figure 5-14 ................... 149

XVI
LIST OF TABLES
Table 3-1: GPS/GNSS Receiver Categories................................................................................. 23
Table 3-2: GNSS Signal Generation Equipment .......................................................................... 30
Table 3-3: GNSS Signals Generated for Test............................................................................... 31
Table 3-4: Minimum Received GNSS Signal Power Levels for Interference Test Events.......... 31
Table 3-5: Test Schedule .............................................................................................................. 41
Table 3-6: Interference Signal Parameters.................................................................................... 42
Table 3-7: CSV Data Format........................................................................................................ 44
Table 3-8: Receivers Tested.......................................................................................................... 58
Table 3-9: Ratio of OOBE limit density to................................................................................... 60
Table 3-10: Wired Test and Data Summary ................................................................................. 61
Table 3-11: Characterized GNSS Antennas ................................................................................. 66
Table 3-12: LNA Performance Characteristics measured with Bench Testing............................ 72
Table 4-1: Base Station Characteristics from ITU-R M.2292...................................................... 74
Table 4-2: Summary of Compiled Use Case Information ............................................................ 77
Table 4-3: Summary of Geometric Parameters ............................................................................ 79
Table 4-4: Coefficients for GLN, GAV, TIM, and HPR Receivers for Modeling Relative VPOL
Antenna Gain at 22 Frequencies................................................................................................... 80
Table 4-5: Coefficients for GLN, GAV, TIM, and HPR Receivers for Modeling Relative HPOL
Antenna Gain at 22 Frequencies................................................................................................... 83
Table 4-6: Tolerable Base Station 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑑𝑑, 𝑓𝑓) for L1 C/A bounding masks for Type-2
Interference signal using FSPL propagation model.................................................................... 102
Table 4-7: Tolerable Base Station 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑑𝑑, 𝑓𝑓) for All GNSS bounding masks for Type-2
Interference signal using FSPL propagation model.................................................................... 103
Table 4-8: Summary Table of Satellite TriG Receiver Characteristics Used for M&S ............. 109
Table 4-9: Zone Model - ES Zone-specific Radial Distance from City Center.......................... 110
Table 4-10: Typical Cell Radius (CR) - ITU-R M.2292 ............................................................ 111
Table 4-11: Total # of ES (Macrocell Deployment Only ........................................................... 111
Table 4-12: Assumed Transmitter Levels per Sector (Typical per ITU-R M.2292) .................. 112
Table 4-13: Summary of Simulation Runs ................................................................................. 113
Table 4-14: COSMIC-2 Interference Results (Macro ES Only, All ES Tx Power 32 dBW)..... 115
Table 4-15: Sentinel-6 Interference Results ............................................................................... 115
Table 5-1: Analysis Scenarios and Conditions........................................................................... 124
Table 5-2: WIRSO Scenario Based Limits from Two Methods................................................. 135
Table 5-3: WIRSO Scenario Based Limits from Random Method............................................ 136
Table 5-4: Comparison of Two Methods for WIRSO Scenario Based Limits........................... 137
Table 5-5: Hexagonal Grid Power Limits Computed Using Two Methods ............................... 141
Table 5-6: Hexagonal Grid Power Limits Computed Using 433m ISD Flat Earth Scenario..... 142
Table 5-7: Power Limits for Landed Helicopter at Various Separation Radii
f rom Central Tower.................................................................................................................... 143
Table 5-8: Hexagonal Grid Scenario Based Limits .................................................................... 144
Table 5-9: Summary of Scenarios & Findings ........................................................................... 151

17
1. INTRODUCTION
The goal of the U.S. Department of Transportation (DOT) Global Positioning System (GPS)
Adjacent Band Compatibility Assessment is to evaluate the maximum transmitted power levels
of adjacent band radiofrequency (RF) systems that can be tolerated by GPS and Global
Navigation Satellite System (GNSS) receivers. The results of this effort advance the
Department’s understanding of the extent to which such adjacent band transmitters impact
GPS/GNSS devices used for transportation safety purposes, among numerous other civil
applications. The assessment described in this report addresses transmitters in bands adjacent to
the 1559-1610 MHz radionavigation satellite service (RNSS) band used for GPS Link 1 (L1)
signals that are centered at 1575.42 MHz.
The assessment had two primary components:
• One component, led by the DOT Office of the Assistant Secretary for Research and
Technology (OST-R), focused on all civilian GPS devices and their applications, apart
from certified aviation. Through this component of the Study, categories of receivers were
evaluated that included aviation (non-certified), cellular, general location/navigation, high
precision, timing, and space-based receivers. An element of this effort was to determine
equipment susceptibility to adjacent band interference to support analyses for deriving
compatible power levels.
• The other component, led by the Federal Aviation Administration (FAA), focused on
certified GPS avionics, and was conducted by analysis to determine the adjacent band
power levels that conform to existing certified GPS aviation equipment standards.
The DOT GPS Adjacent Band Study is the product of an extensive process to gather stakeholder
views and input. OST-R and FAA benefited significantly from feedback received via
governmental and public outreach. This feedback was important to ensure broad agreement and
understanding of equipment use cases, interaction scenarios, propagation models, and transmitter
characteristics.
For the OST-R component of the effort, the first public workshop was held in September 2014 at
DOT’s Volpe Center in Cambridge, MA. Five subsequent workshops were held at locations on
both coasts of the United States (Los Angeles, CA and Washington, DC) to obtain broad
stakeholder participation. These workshops presented the elements of the OST-R assessment:
equipment susceptibility testing, development of use cases and interaction scenarios, transmitter
characteristics, and propagation modeling assumptions, and finally, the analysis and assessment
results. Initial planning of the DOT GPS Adjacent Band Compatibility Assessment focused on
testing receivers that only process GPS signals. However, based upon feedback from public
outreach, the assessment was expanded to include widely available equipment that also processes
GNSS signals from other satellite navigation constellations in the 1559-1610 MHz band.

18
While the compatibility assessment is intended to be generally applicable in terms of the type of
adjacent band system, the main focus for this L1 band assessment was on Long Term Evolution
(LTE) signals. The OST-R effort included extensive equipment testing to derive interference
tolerance masks (ITMs). The ITM defines, for a given frequency, the maximum power allowed
to ensure the tested GPS/GNSS receiver did not experience more than a 1 dB reduction in
carrier-to-noise density ratio (CNR) for various categories of GPS/GNSS receivers. The receiver
ITMs were derived from radiofrequency equipment testing, both radiated and conducted, for
frequencies ranging from 1475 MHz to 1675 MHz (GPS L1 +/- 100 MHz). These ITMs were
then used with appropriate use cases and interaction scenarios to determine maximum transmitter
EIRP levels that could be tolerated from adjacent band transmitters.
The equipment susceptibility testing involved 80 GPS/GNSS receivers tested in an anechoic
chamber in April 2016. The GPS/GNSS receivers for this test effort were provided by U.S.
Government (USG) partners and industry and represented the diverse nature of GPS/GNSS
applications and services. In addition to this primary test effort, more focused testing on a subset
of equipment was conducted with wired testing in a laboratory setting and antenna
characterizations in a different anechoic chamber. The receiver test data from the primary test
effort was analyzed to develop ITMs, based on a 1 dB CNR degradation, which provided
bounding performance for each GPS/GNSS receiver category.
These bounding ITMs and GPS/GNSS antenna characteristics were the primary inputs to use
case scenario assessments to determine the maximum Effective Isotropic Radiated Power (EIRP)
that could be tolerated in the adjacent radiofrequency bands for each GPS/GNSS receiver
category. Space-based applications are different from other GPS/GNSS applications considered,
primarily due to the need to account for aggregation effects of multiple transmitters visible in
orbit. Although OST-R derived ITMs for space-based receivers, along with other GPS/GNSS
receiver types, OST-R deferred to the National Aeronautics and Space Administration (NASA)
for assessing adjacent-band transmitter power levels that can be tolerated for this receiver
category.
The FAA’s public outreach for their component of the GPS Adjacent Band Compatibility
Assessment was initiated in early 2014 with RTCA Inc., an aviation advisory body. This
outreach was followed in October 2014 with a document detailing the FAA’s approach to the
assessment for certified aviation and the request to RTCA to vet assumptions and respond to
specific questions. These questions ranged from receiver/antenna characteristics and their
applicability to fixed- and rotary-winged aircraft to specific propagation modes to be used and
interaction scenarios.
RTCA also was requested to comment on use of an exclusion zone concept and its implications
for operations and flight safety. RTCA completed the review and provided comments to FAA in
2015. The FAA was approached starting in 2016 by one entrant with an analysis approach for
certified aviation that included a specific transmitter network and exclusion zone. This proposal

19
was reviewed by RTCA and was considered with the material originally vetted by them in
FAA’s assessment of maximum tolerable EIRP for certified aviation.
The FAA effort did not require receiver and antenna equipment testing because certified aviation
receiver standards specify the maximum tolerable interference environment to ensure all receiver
functions are protected. The FAA effort also considered use cases based upon one specific,
proposed adjacent-band LTE network.
This report is organized as follows. Section 2 provides background information to the study.
Section 3 discusses the radiated and wired tests performed and provides results for all civil
receiver categories with the exception of the certified aviation receiver category. Section 4
presents the analysis to determine the tolerable transmit power levels, including use cases for
applications other than certified aviation. Analysis of aggregate effects for on-orbit space
applications is provided by the National Aeronautics and Space Administration (NASA).
Section 5 discusses the analysis approach and presents the results for the certified aviation
receiver category. Section 6 provides an overall summary of the report. Additional information
on test results and analyses are included in the appendices.

20
2. BACKGROUND
Over the past three decades, GPS has grown into a global utility providing multi-use service
integral to U.S. national security, economic growth, transportation safety, and homeland security,
and as an essential element of the worldwide economic infrastructure. GPS affects the lives of
the American public every day, ranging from its use in all modes of transportation to
incorporation of GPS timing into the electric grid, communications networks, point of sale
transactions, banking and finance, as well as applications of GPS for surveying, precision
agriculture, weather forecasting, earthquake monitoring, and emergency response. The range of
commercial and civil applications of GPS continues to expand and the importance of many GPS
and GNSS applications has significantly increased.
Private sector innovations in the use of GPS greatly exceed any originally envisioned or
imagined applications. However, unlike communication systems where performance
improvements are enabled by coordinated changes to both the transmitting and receiving
systems, GPS has shown that user processing innovations can significantly improve performance
without changing the transmitted GPS signals. These innovations have enabled the civil
community to develop and implement new GPS antenna/receiver technologies and applications,
with minimal dependency on government actions. As the economic and security importance of
positioning, navigation, and timing (PNT) gained international recognition, other countries have
initiated or renewed their commitments to provide satellite navigation systems, fueling further
development of new user-based GPS/GNSS technologies.
The framework for GPS policy is defined by Presidential Policy. Title 10 United States Code,
Section 2281 (b) states that the GPS Standard Positioning Service shall be provided for peaceful
civil, commercial and scientific uses on a continuous worldwide basis. The 2010 National Space
Policy sustains the overall radiofrequency environment in which critical U.S. space systems such
as GPS operate and calls for continued U.S. leadership in the service, provision, and use of
GNSS.
This policy reaffirms existing U.S. commitments under National Security Presidential Directive
(NSPD)-39, U.S. Space-Based Positioning, Navigation, and Timing Policy (15 December 2004)
to provide continuous, worldwide access to civil GPS, free of direct user fees; pursue
international GNSS cooperation including use of foreign PNT to augment and strengthen the
resiliency of GPS; operate and maintain GPS to meet published standards; and take steps to
detect and mitigate GPS interference. Per NSPD-39, DOT serves as the civil lead for GPS.
At the direction of the DOT Deputy Secretary, FAA and OST-R developed the GPS Adjacent
Band Compatibility Assessment Plan to provide a means to advance the Department’s
understanding of the adjacent radiofrequency band power levels that would be compatible for
GPS civil applications. The plan identifies the processes to: (a) derive adjacent-band transmitter
power limit criteria for assumed new applications necessary to ensure continued operation of
GPS services, and (b) determine similar levels for future GPS receivers utilizing modernized
GPS and interoperable GNSS signals [1]. This document provided the framework for the

21
processes and assumptions that resulted in the testing and analysis conducted during the effort
and presented in this report.

22
3. CIVIL RECEIVER TESTING
3.1 Anechoic/Radiated Testing
In planning and preparation for receiver testing, OST-R held multiple public workshops to
discuss plans for the Study and to foster the exchange of information among interested parties.
These workshops took place on September 18, 2014 (see 79 Fed. Reg. 47171), December 4,
2014 (see 79 Fed. Reg. 68345), March 12, 2015 (see 80 Fed. Reg. 8125), and October 2, 2015
(see 80 Fed. Reg. 57915). Representatives from NTIA, FCC, and NIST, and other Space-Based
PNT EXCOM departments and agencies participated in the public workshops.
A draft test plan was issued for public comment on September 9, 2015 (see 80 Fed. Reg. 54368).
There were six organizations and individuals who provided written comments on the draft of the
test plan: Ligado, GPS Innovation Alliance, Greenwood Telecom, Alliance for
Telecommunications Industry Solutions (ATIS), General Motors, and Logan Scott. The
Department carefully reviewed and considered the comments that were submitted in devising a
final test plan, as well as other information that was offered in the course of the public
workshops. In addition, DOT made public its Nondisclosure Agreement (NDA) for the
protection of certain confidential or proprietary information that may be offered by companies
that participated in the Study (see 81 Fed. Reg. 12564). DOT executed five NDAs with Deere &
Company, GM Global Technology Operations LLC, Novatel Inc., Trimble Navigation Limited,
and u-Blox AG.
After the Test Plan was finalized and published (see 81 Fed. Reg. 12564), GPS/GNSS receiver
testing, led by the OST-R/Volpe Center, was conducted at the U.S. Army Research Laboratory
(ARL) at the White Sands Missile Range (WSMR) facility in New Mexico in April of 2016.
Results from the testing described in this section were presented at public workshops held on
October 14, 2016 (see 81 Fed. Reg. 68105) and on March 30, 2017 (see 82 Fed. Reg. 13924).
Information from all of the public workshops that were held can be found at a website hosted by
the National Space-Based PNT Coordination Office (NCO) at
http://www.gps.gov/spectrum/ABC/.
Devices Under Test (DUTs)
DOT sought to include a broad range of devices used in rail, aviation, motor vehicle, maritime,
and space applications, among a number of other civil uses of GPS/GNSS including timing,
surveying, precision agriculture, weather forecasting, earthquake monitoring, and emergency
response. The GPS/GNSS receivers for this test effort were provided by USG partners and
industry.
Six categories of GPS/GNSS receivers were considered for the OST-R portion of the effort,
which are identified in Table 3-1. High precision (HPR) and the differential Network (NET)
receivers are grouped together into one category since HPR receivers are commonly used in
differential networks. General aviation receivers include non-certified receivers and are separate
from certified aviation receivers which did not require testing since existing certified aviation

23
receiver standards specify the maximum tolerable interference environment to ensure all receiver
functions are protected. Space-based receivers were included with assistance from NASA.
Table 3-1: GPS/GNSS Receiver Categories
Number Category Abbreviation
1
General Aviation (Non-
Certified)
GAV
2 General Location/Navigation GLN
3 High Precision/Networks HPR/NET
4 Timing TIM
5 Cellular CEL
6 Space Based SPB
GPS/GNSS Receivers Tested
During the WSMR anechoic radiated chamber testing in April 2016, DOT and other participants
tested 80 GPS/GNSS receivers listed along with the associated antennas in Table 3-2.
Duplicated entries in Table 3-2 indicate that two identical receiver/antenna model pairings were
tested, which occurred in three instances. In addition, 14 (out of the 80) were subsequently
subjected to additional conducted/wired testing at Zeta Associates as indicated by an asterisk
next to the receiver name in Table 3-2. The Air Force GPS Directorate conducted testing of
military GPS receivers the week prior to the civil receivers being tested.
Table 3-2: List of GPS/GNSS Receivers Tested at WSMR
Receiver Antenna
Android S5* Integrated
Android S6 Integrated
Android S7 Integrated
Arbiter Systems 1088B-Satellite Control
Clock
Arbiter AS0087800
Arbiter Systems 1094B-GPS Substation
Clock
Arbiter AS0087800
Ashtech uZ-CGRS Choke Ring
Ashtech Z-12 Choke Ring
Dual Electronics - SkyPro XGPS 150 Integrated
Dynon 2020 Integrated
Dynon 250 Integrated
EVA-7M EVK-7EVA-0
AeroAntenna Technology Inc AT2775-41-
TNCF

24
EVA-M8M EVK-M8EVA-0
TNCF
EVK-6n Passive patch
EVK-7P Passive patch
EVK-M8N* Passive patch
EVK-M8T Passive patch
Furuno GP-33 GPA017/19
Garmin - Area 560
TNCF
Garmin - GLOGPS (GPS & GLONASS) Integrated
Garmin - GPSMap 696*
TNCF
Garmin EDGE 1000 Integrated
Garmin ETREX 20x Integrated
Garmin GPSMap 295
TNCF
Garmin GPSMAP 64 Integrated
Garmin GPSMAP 741 Garmin GA 38 GPS/GLONASS antenna
Hemisphere R330 Hemisphere A42
Javad Delta II* JAVAD JAVRINGANT_DM
Javad Delta-3 Choke Ring
Javad EGGDT-160 Choke Ring
JAVAD Triumph-1 Integrated
LEA-M8F EVK-M8F-0
TNCF
LEA-M8S EVK-M8N-0
TNCF
Leica GR10 TRM59800.00
Leica GRX1200GGPRO Leica AX1202GG
Leica GRX1200GGPRO* LEIAT504
MAX-7C EVK-7C-0
TNCF
MAX-7Q EVK-7N-0
TNCF
MAX-M8Q Passive patch
MAX-M8Q EVK-M8N-0
TNCF
NAVCOM SF3050* NAVCOM ANT-3001R
NovAtel 628 Card w/ Flex pack 703GG
Novatel OEM628V-G1S-B0G-TTN-H
installed in Development board
Patch
Schweitzer Eng. Labs SEL-2401-Satellite
Synchronized Clock
SEL 235-0209
Septentrio PolaRx4Pro* AERO AERAT1675_120
Septentrio PolaRx4TR Pro Choke Ring

25
Septentrio PolaRx5TR Pro* TRM59800.00
Septentrio PolaRx5TR Pro* TRM59800.00
SF3000 Integrated
SF3000 Integrated
SiRF III
TNCF
Supercruise "VCP" Shark Fin
Supercruise "VCP" Shark Fin
Symmetricom SyncServer S350 AeroAntenna AT575-142
Symmetricom Xli Symmetricom Antenna 1
Symmetricom Xli AeroAntenna AT575-142
Symmetricom-GPS Symmetricom Antenna 2
Topcon Net-G3A Sigma Topcon CR-G3
TriG Choke Ring
TriG V2 Choke Ring
Trimble 5700* Trimble TRM41249.00
Trimble Acutime 360 Integrated
Trimble Ag-382 Integrated
Trimble Ag-382 Integrated
Trimble Bison III Trimble 70229-52
Trimble Geo 7X Integrated
Trimble NETR5 Trimble TRM55971.00
Trimble NETR5 Trimble Zephyr Geodetic Model 2
Trimble NETR5* Trimble Zephyr 59800-00
Trimble NETR9 TRM59800.00
Trimble NETR9* Trimble TRM29659.00
Trimble NETRS Ant com Active L1/L2
Trimble NETRS Trimble Zephyr Geodetic
Trimble NetRS TRM59800.00
Trimble NETRS* Ashtech ASH701945B_M
Trimble R8 Integrated
Trimble SMT360 GPS receiver* Trimble SMT-360 Antenna
Trimble SPS461 GA530 Ruggedized
Trimble SPS855 Trimble Zephyr 2
Trimble SPS985 Integrated
uBlox EVU-6P-0-001
TNCF
*Subsequently subjected to additional conducted/wired testing at Zeta Associates
Antennas Tested
In addition to the antennas listed in Table 3-11, that were subjected to radiated testing at WSMR
while connected to the corresponding receivers listed in this table, a subset of these antennas and
some additional antennas were subsequently characterized in a smaller anechoic chamber at
MITRE in Bedford, MA. See Section 3.3.1.

26
Anechoic Chamber
The radiated adjacent band testing was performed at the Army Research Laboratory (ARL)
Anechoic Chamber located at WSMR. The test configuration and approximate dimensions for the
Electromagnetic Vulnerability Assessment Facility (EMVAF) are shown in Figure 3-1.
The GNSS equipment test area was approximately 24’ × 24’ and was radiated from above using
two separate antennas. One antenna radiated the interference signals while the other radiated
GNSS signals with both approximately 25’ above the center of the test area. The signal
generation equipment was located on the mezzanine platform while participant collection and
support equipment was located at the opposite end of the chamber. To emulate standard field
operation of each particular receiver as closely as possible, some receivers were located in the
participant area and RF cables were run to their respective antennas. Receivers with integrated
antennas were placed directly in the test area with data collection/control cables typically routed
to the participant area.
Figure 3-1: Chamber Dimensions and Layout
Location in Chamber
During the radiated testing, there were 12 participating organizations including DOT’s Federal
partners and agencies, and GPS/GNSS receiver manufacturers. The organizations included: 1)
United States Coast Guard (USCG), 2) NASA, 3) National Oceanic and Atmospheric
Administration (NOAA), 4) United States Geological Survey (USGS), 5) FAA, 6) U.S. DOT, 7)
General Motors (GM), 8) u-blox, 9) NovAtel, 10) Trimble, 11) John Deere, and 12) UNAVCO, a

27
non-profit university-governed consortium that facilitates geoscience research and education
using geodesy sponsored by NASA and the National Science Foundation. The participating
organizations, number of receivers and location on the test grid can be found in Figure 3-2. A
cavity backed spiral antenna used for signal calibration and chamber mapping was placed at the
edge of the test grid between locations E0 and G0. This antenna was connected to a spectrum
analyzer and used for continuous signal and interference monitoring. In addition, a horn antenna
connected to a spectrum analyzer was setup in the middle of the test grid (E7) for signal and
interference monitoring and situational awareness for participants.
Figure 3-2: DUT and Calibration Grid Locations
←Exit
K0 K2 K4 K6 K8 K10 K12
K1 K3 K5 K7 K9 K11 K13
J6
J5
I0 I2 I4 I6 I8 I10 I12
I1 I3 I5 I7 I9 I11 I13
H8
H7
G0 G2 G4 G6 G8 G10 G12
G1 G3 G5 G7 G9 G11 G13
E0 E2 E4 E6 E8 E10 E12
E1 E3 E5 E7 E9 E11 E13
D8
D7
C0 C2 C4 C6 C8 C10 C12
C1 C3 C5 C7 C9 C11 C13
B6
B5
A0 A2 A4 A6 A8 A10 A12
A1 A3 A5 A7 A9 A11 A13
Note: not drawn to scale
Mezzanine
Calibration Test Locations
GM
16
UNAVCO
5
12
11
2
4
3
USGS TRIMBLE DEERE
8
10
6
2
NOVATEL
1
5
9
8
UBLOX
7 13
Aero_Spec_A
1
8
2
1 2
USDOT
USCG NASA NOAA
3
9
5
FAA
23

28
Figure 3-3 is a photo of the DUTs during the anechoic chamber testing.
Figure 3-3: Photo of the DUTs on the test grid
GNSS Signal Generation
The GNSS signal generation and recording process is shown in Figure 3-4. The recording was
conducted at MITRE prior to the test period. Simulated satellite signals were generated using a
set of four Spirent GSS8000 GNSS signal simulators. These are commercial research and test
devices that produce high-fidelity RF signals as they would appear at the output of a GNSS
receive antenna. The GSS8000 simulators allow specification of received signals, received
signal power level, satellite orbital parameters, user location, etc. The simulators were
programmed to synchronously generated signals for GPS+ Wide Area Augmentation System
(WAAS), Beidou, GLONASS, and Galileo. The L1+L2 radio frequency outputs of the
simulators generating the GPS+WAAS, Beidou, and Galileo signals were passively combined
using a single channel of a Spirent GSS8368 Signal Combiner. The GLONASS L1 signals were
passed through the second channel of the GSS8368 signal combiner.

29
Figure 3-4: GNSS Signal Generation and Recording
The resultant RF data was recorded using a set of three National Instruments (NI) PXIe-5663E
Vector Signal Analyzers housed in a National Instruments PXIe-1075 chassis. The three
channels were recorded into 26.4 MHz wide bands centered at 1227.6 MHz, 1572.2 MHz, and
1602 MHz at 33 MS/s using 16 bit complex samples. The sampled data was then recorded on an
NI HDD-8265 12 Terabyte Redundant Array of Independent Disks (RAID) (see Figure 3-4).
The total recording time for the interference test events is approximately seven hours in length.
The GNSS signal playback and transmission process is illustrated in Figure 3-5. During the test
period, the recorded GNSS signals were converted back to RF using a set of three National
Instruments PXIe-5673E Vector Signal Generators (VSG) using the same sample rate as was
used to record the data. The outputs of the three signal generators were combined using a Narda
model 4372A-3 passive 3-port combiner. The output of this combiner was passed through a
Vaunix Technology LDA-602 variable attenuator.
The attenuation level was established during chamber mapping and calibration (see Section
3.1.4) such that the received signal strength was at or above the power levels specified below.
The output of the attenuator was followed by a splitter (that was connected to a spectrum
analyzer during the test) and then an RF isolator with approximately 60 dB of isolation to
prevent RF power from entering the system through the antenna. The isolator was connected to
a custom passive L1/G1/L2 GNSS patch antenna that was suspended from the ceiling of the test
chamber (see Figure 3-1). Prior to the WSMR deployment, the GNSS signal playback system
was tested in a chamber at MITRE to ensure that the playback accurately reproduced the desired
scenarios.

30
Figure 3-5: GNSS Signal Playback and Transmission
The equipment that was transported to and set up in the chamber is listed in Table 3-2.
Table 3-2: GNSS Signal Generation Equipment
Equipment Make/Model Notes
RAID storage National Instruments HDD-8265 12 TB
3-channel VSG chassis National Instruments PXIe-1075 Includes computer controller
(PXIe-8133) with LabView
software.
VSGs (3 each) National Instruments PXIe-
5673E
Each VSG consists of NI PXIe-
5450 (400 MS/s I/Q Signal
Generator), PXIe-5611 (I/Q
Vector Modulator), and PXIe-
5652 (RF Signal Generator).
Combiner Narda 4372A-3 Passive 3-port
Digitally-controlled variable
attenuator
Vaunix LDA-602 Provides up to 50 dB
attenuation; controlled by VSG
chassis computer.
Isolator Addington Laboratories 222-
0170A
Provides approximately 60 dB
isolation.
Passive GNSS antenna MITRE custom RHCP antenna covers 1559 –
1610 MHz and 1212 – 1242
MHz
Rb Frequency Source Symmetricom 8040 10 MHz
Amplifiers MiniCircuits ZRL-2400-LN 23-30 dB gain. Used for
chamber calibration (see Section
4.3.4.2).
Cables Various As needed and with appropriate
connectors.

31
The GNSS signals that were generated and recorded at MITRE and then broadcast in the
chamber at WSMR are indicated in Table 3-3.
Table 3-3: GNSS Signals Generated for Test
Signal
GPS L1 C/A-code
GPS L1 P-code
GPS L1C
GPS L1 M-code
GPS L2 P-code
SBAS L1
GLONASS L1 C
GLONASS L1 P
BeiDou B1I
Galileo E1 B/C
For the interference test events, the transmitted GNSS signal powers were calibrated to yield the
minimum signal levels specified in Table 3-4 at the location in the test grid with the lowest
received power (see Section 3.1.4). The signal powers were held constant over the duration of
the event, except for the linearity test.
Table 3-4: Minimum Received GNSS Signal Power Levels for Interference Test Events
Signal Minimum Received Power out of 0 dBic antenna (dBW)
GPS C/A-code -158.5 for 8 SVs, -168.5 for 1 SV, -178.5 for 1 SV
GPS L1 P(Y)-code -161.5 for 8 SVs, -171.5 for 1 SV, -181.5 dBW for 1 SV
GPS L1C -157 for 8 SVs, -167 for 1 SV, -177 for 1 SV
GPS L1 M-code -158 for 8 SVs, -168 dBW for 1 SV, -178 dBW for 1 SV
GPS L2 P(Y)-code -164.5 for 8 SVs, -174.5 for 1 SV, -184.5 for 1 SV
GPS L2 M-code -161 dBW for 8 SVs, -171 dBW for 1 SV, -181 dBW for 1 SV
SBAS L1 -158.5 for 2 SVs
GLONASS L1 C -161 for 10 SVs, -171 for 1 SV, -181 for 1 SV
GLONASS L1 P -161 for 10 SVs, -171 for 1 SV, -181 for 1 SV
BeiDou B1I -163 for 10 SVs, -173 for 1 SV, -183 for 1 SV
Galileo E1 B/C -157 for 10 SVs, -167 for 1 SV, -177 for 1 SV
The user was located at 32N, 106W, 0 m height relative to the WGS-84 ellipsoid and was
stationary over the simulated time span. The simulated start time was April 18, 2016 08:00

32
MDT. Yuma-style almanacs for the GPS, GLONASS, BeiDou, and Galileo constellations are
provided in Appendix A.
Errors that the GNSS simulators were capable of emulating were set to zero except for
ionospheric delay and tropospheric delay errors, which are described in Appendix A. The
objective of the simulator configuration was to minimize pseudorange and carrier phase errors
from all sources (e.g., satellite clock errors, satellite ephemeris errors, residual ionosphere,
residual troposphere) for the devices under test to enable measurement of the introduced
interference source effects without the influence of other errors that are not attributable to the
interference source. Since GNSS receivers typically apply tropospheric and ionospheric
correction models, these error sources were emulated to minimize residual receiver measurement
errors.
Interference Signal Generation
The Software Programmable Interference Generator for ABC Testing (SPIGAT) was assembled
to generate additive white Gaussian noise (AWGN) and simulated LTE signals as interference at
controlled power levels at specified frequencies. This system was automated to execute these
tests for a suite of 22 discrete interference frequencies at appropriate signal levels with minimal
operator intervention. The frequencies and signals levels tested are shown in Figure 3-6 and the
interference system configuration for the radiated test is depicted in Figure 3-7. The VSG
generated either 1 MHz AWGN (Type 1) or LTE (Type 2) signals at a fixed level at the
appropriate carrier frequency and the adjustable attenuator controlled the signal level input from
the VSG to the high power amplifier (HPA).
Using the attenuator in this fashion ensured Adjacent Channel Leakage Ratio (ACLR) from the
VSG was maintained through different test levels. The HPA output was provided to an RF
switch that directed the interference signal through one of 22 RF cavity filters. Lastly, the
amplified and filtered interference signal was directed to a linearly polarized standard gain horn
transmit antenna that irradiated the GNSS receivers under test. Directional couplers were
included in the signal generation path to provide monitor points and a circulator was added to
provide overload protection.

33
Figure 3-6: Interference Frequencies and Signal Levels Tested
Figure 3-7: Interference System Configuration for Radiated Test
Test execution for each interference test signal was controlled by a pre-defined configuration
file. The configuration file contained the frequencies to be tested, power levels, signal type, and
test durations. The control computer was “GPS time aware” by obtaining receiver time from a
GPS receiver tracking signals directly from the GNSS signal generator. This allowed time
alignment with receiver data during processing. The interference control software recorded GPS
time and instrument settings once per second into SPIGAT summary files for each test event.

34
One of the more important considerations for SPIGAT was the RF filters and ensuring they were
sufficient to attenuate out-of-band emissions (OOBE) so that degradation measured was due
almost entirely to the fundamental signal only. (The impacts of OOBE with the fundamental
signal were investigated in the wired testing.) The RF passband cavity filters used to condition
the interference were grouped into two categories based on roll-off specification. Filters
centered at test frequencies nearer (but outside of) the RNSS band, namely 1540, 1545, 1550,
1555, 1615, 1620, and 1625 MHz were designed to meet tighter roll-off requirements with the
primary feature being 65 dB rejection at 9 MHz from center. The remaining filters were
designed to meet a more relaxed set of requirements with 65 dB rejection at 20 MHz from center.
The measured frequency dependent gain for these two filter types are shown in Figure 3-8.
Further details on SPIGAT and WSMR test conditions are provided in Appendix A.
Figure 3-8: Measured gain response for (a) bandpass RF filter with tighter rejection
requirements, (b) bandpass RF filter with more relaxed rejection requirements.
Type 1 Signals
The Type 1 signal was selected for testing at all 22 frequencies to provide a narrowband signal
for assessment of a general receiver mask not specifically tied to LTE. The signal tested was
bandpass AWGN with bandwidth B=1 MHz. Some early consideration was given to using
continuous wave (CW) interference but this 1 MHz signal was adopted based primarily on
concerns some receivers might employ techniques specifically designed to mitigate CW signals.
The AWGN signal was generated offline and had a duration of approximately 0.021 seconds.
This signal file was loaded into the VSG, up-converted to the appropriate frequency, and played
out continuously end to end from VSG memory. Figure 3-9 is a spectrum analyzer capture of the
Type 1 signal as received in the EMVAF test area during testing.
(a) (b)

35
Figure 3-9: Type 1 Signal Captured during WSMR Testing @ 1530 MHz
Type 2 Signals
The Type 2 signal emulated LTE characteristics representing both a downlink and an uplink. The
downlink was emulated as a fully loaded base-station with Orthogonal Frequency Division
Multiplexing (OFDM), and the uplink with Sub-Carrier OFDM (SC-OFDM). The LTE digital
waveforms were generated using the MATLAB LTE package. As with the Type 1 signal, these
LTE representations were loaded on the VSG, up-converted to the appropriate carrier frequency,
and played out end to end from VSG memory. The durations of the downlink and uplink files
were two seconds each. The specific settings and commands used in MATLAB are provided
with Appendix A. Figure 3-10 is a spectrum analyzer capture of the Type 2 signal as received in
the EMVAF test area during testing.

36
Figure 3-10: Type 2 Signal Captured during WSMR Testing @ 1530 MHz
Intermodulation
The Type 1 and 2 interference signals provide a measure of the effect of an interfering signal on
a particular GPS/GNSS receiver but do not capture impacts of spurious emissions due to two or
more signals operating simultaneously at different center frequencies. The intermodulation
signal test was included to demonstrate this potential impact by simultaneously transmitting
Type 2 downlink signals at center frequencies of 1530 MHz and 1550 MHz. The 3rd
order
intermodulation product of these center frequencies falls near the center of the L1 band. For this
specific test, the 1550 MHz signal was generated so that it was approximately 10 dB lower than
the 1530 MHz signal. The downlink LTE representation discussed earlier was up-converted to
1530 and 1550 MHz and played out end to end from VSG memory. The duration of the
intermodulation file was 0.5 seconds. Figure 3-11 is a spectrum analyzer capture of the
intermodulation signal as received in the EMVAF test area during testing.

37
Figure 3-11: Intermodulation Signal Captured during WSMR Testing
System Calibration and Chamber Mapping
System calibration and chamber mapping included several efforts at WSMR; 1) GNSS
calibration, 2) SPIGAT calibration, 3) and, chamber mapping. GNSS calibration is described in
Section 3.1.4.1 and the remaining topics are discussed in Section 3.1.4.2 and Appendix A.
GNSS System
To calibrate the GNSS signal power levels, the playback system was modified slightly from the
configuration shown in Figure 3-5. Two LNAs were inserted after the variable attenuator to
increase the output power level. This power increase allowed the received GNSS signal levels to
be accurately measured by a spectrum analyzer connected to a RHCP cavity-backed spiral
antenna that was moved across 45 points in the test grid. The gains of the two LNAs were
determined from measurement (see Figure 3-12).
The calibration proceeded in two steps. First, the VSGs in the playback system were utilized to
produce tones at two frequencies (1227 MHz and 1561 MHz) to determine the variation in
received power across the test grid at these frequencies. These variations, as measured using a
RHCP cavity-backed spiral antenna at 45 locations spanning the test grid, are shown in Figure
3-13 and Figure 3-14. The numerical values on each of these plots are in dBm.
Second, to establish the output power of the three VSGs in the playback system, a GNSS signal
from a single satellite was emulated (GPS P-code for 1227 MHz, GPS P-code at 1575 MHz, and

38
GLONASS C for 1602 MHz) and the received power measured at the location at the edge of the
test grid where the received power was a minimum. The measured power, adjusted by the known
LNA and cavity-backed spiral antennas gains, was compared to the target received power levels
in Table 3-4. The resultant differences were used to establish power settings for each of the three
VSGs in the playback system for the test events. The VSG output powers were adjustable both
relatively and absolutely (through power commands sent digitally by computer interface to each
VSG to change relative gains, and through removal of the two LNAs as well as commands to the
digital variable attenuator that followed the combined VSG outputs). The objective of this
calibration process was to ensure that the power levels specified in Table 3-4 were achieved or
exceeded out of a 0 dBic receive antenna at any location in the test grid. As evident from Figure
3-13 and Figure 3-14, when the target minimum received GNSS signal levels were achieved at
the worst-case location in the grid (grid corners), they were exceeded by up to 3.7 dB at the
center of the grid.
Figure 3-12: Measured Gain of Two MiniCircuits ZRL-2400-LN Amplifiers
1200 1250 1300 1350 1400 1450 1500 1550 1600 1650
23
24
25
26
27
28
29
30
Freq. (MHz)
Gain
(dB)
X: 1228
Y: 27.99
X: 1575
Y: 24.12
X: 1602
Y: 26.08
X: 1560
Y: 24.32
AMP10
AMP11

39
Figure 3-13: GNSS Signal Received Power Variation across the Test Grid (1227 MHz)
Figure 3-14: GNSS Signal Received Power Variation across the Test Grid (1561 MHz)

40
SPIGAT
The interference system was calibrated to determine the system biases for each interference
frequency and to ensure the intended power levels were achieved in the test area. This
calibration was accomplished by selecting a reference location at the edge of test area at
approximately the 3 dB beamwidth of the interference transmit antenna and measuring CW tones
generated by SPIGAT at all 22 discrete frequencies. The CW tones were measured using a
cavity backed spiral antenna (AST-1507AA) mounted on a tripod, calibrated RF cable, and a
spectrum analyzer (Agilent E4445A). This procedure resulted in a correction table per frequency
utilized by SPIGAT. The interference linearity was also measured from this reference location
to demonstrate received power over the test range at each frequency matched the intended levels.
3.1.4.3 Chamber Mapping
Chamber mapping was conducted after calibration was complete to determine RF power
variation across the test area. Mapping used a grid of 45 measurement points separated by
approximately four feet which encompassed the test area. Chamber mapping was completed at
the beginning and end of DOT testing with all equipment installed in the test area. These two
mappings were very consistent and final mapping values used with receiver processing
represented their average. Examples of the final power mapping at 1475, 1575 and 1675 MHz
are provided in Figure 3-15. This figure shows the expected performance of the interference
antenna beamwidth becoming narrower at higher frequencies. The location of each receiver
tested was known relative to the mapping grid and this mapping data was used to modify
SPIGAT test event summary files and generate unique interference power profiles per frequency
for each receiver location.
Figure 3-15: Power Correction Representations for Three Frequencies
Test Sequence
The test schedule executed at WSMR is shown in Table 3-5. In addition to Type 1, Type 2, and
intermodulation signals described previously there was also a CNR linearity test. Table 3-5
shows the day each test was executed and test number. Type 1 and Type 2 signal tests were

41
given priority so these were tested on separate days to increase the likelihood of obtaining a more
robust data set. The in-band test mentioned in this table was the result of removing Type 1
signals directly in the RNSS band from this test sequence and testing separately. A discussion of
the rationale for this change is provided later in this section. Lastly, each test event listed in the
table was preceded by a stabilization period of at least fifteen minutes with the GNSS system
turned ON to allow participants to reset equipment, verify GNSS signal tracking, and ensure data
collection was started.
Table 3-5: Test Schedule
Day of Week (24-28 April 2016) Test and Number
Monday CNR linearity - Test01
Tuesday Type 1 signal - Test02
Type 2 signal – Test03
Wednesday Type 2 Signal – Test04
Type 1 Signal – Test05
Thursday In-Band – Test56
CNR linearity - Test07
Intermodulation – Test89
Linearity Test
The GPS/GNSS receivers’ CNR estimators need to operate in their linear region. A linearity test
was conducted on the CNR estimators varying GNSS signal power. SPIGAT was not used for
this test event. For this test, GNSS signals of each type were generated to match their intended
levels during interference testing (i.e. majority of GPS L1 C/A were generated at-158.5 dBW and
two SVs were at the lower specified power levels of -10 and -20 dB). After five minutes at these
nominal levels, the test sequence had each signal power increased by two dB every 15 seconds
until they reached +10 dB relative to the nominal level. Each signal power was then decreased
by two dB every 15 seconds until they reached -20 dB relative to the nominal level. Finally, each
signal power was increased by two dB every 15 seconds until power was returned to nominal
levels. The receivers were allowed to track for a brief period following the last signal power step
before concluding this test event.
Interference Test
Each of the individual interference tests used the exact same sequence. The interference test
sequence for each frequency started with a quiescent period of five minutes with no interference
to establish baseline CNR followed by stepping through the full power range at two dB steps
with 15 second dwells at each level. This sequence was repeated for all desired frequencies for
each interference signal. Table 3-6 shows the specific frequencies, power levels, and LTE signal
types used in this testing. The power range for the intermodulation test event was -90 to -20 dBm
for the signal generated at 1530 MHz and -100 to -30 dBm for 1550 MHz.

42
As mentioned above, the Type 1 test event had the two in-band frequencies (1575 MHz and 1595
MHz) extracted from the full set of 22 test frequencies and made into a separate test event.
During the set-up period at the chamber, some system verification testing revealed that for these
two in-band frequencies noise from the high-power amplifier (VSG power turned OFF) would
affect receiver performance. The system reconfiguration required to circumvent this issue was to
place a 20-dB attenuator at the output of the interference generation system which effectively
lowered the output noise floor. The interference power was increased for the addition of this 20-
dB attenuator to meet the desired interference test range. For in-band and intermodulation test
events, SPIGAT was commanded to run two interference test cycles back to back (hence, Test56
and Test89 designations).
Table 3-6: Interference Signal Parameters
𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂𝐂 𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅𝐅
(MHz)
[𝒑𝒑𝒎𝒎𝒎𝒎𝒎𝒎,𝒑𝒑𝒎𝒎𝒎𝒎𝒎𝒎]
(dBm)
LTE Interference Signal Bandpass
Noise
Interference
Signal
Signal Bandwidth
LTE
Type
1475 [-80,-10] 10 MHz, LTE Downlink 1 MHz
1555 [-100,-30] - - 1 MHz
1575 [-130,-60] - - 1 MHz
1595 [-130,-60] - - 1 MHz
1615 [-100,-30] - - 1 MHz
1620 [-100,-30] 10 MHz, LTE Uplink 1 MHz
Data Processing/ITM formation
During testing, organizations used their own programs/software for data collection. This
delegation of data collection responsibility was necessary since many of the 80 receivers had
proprietary interfaces. At the end of each test day, data collected from each receiver was
transferred to DOT’s master data repository. The participants were requested to provide a data
acquisition system (e.g. laptop) with DVD/CD recording capability or asked to use USB hard
drives to transfer data. All data was archived prior to the receivers and participants leaving the
test area.

43
Data Conversion and Format
The master data repository was setup inside the chamber to allow participants to transfer data
from each receiver to this repository. The master repository consisted of a desktop with a local
storage array which accommodated all the data. Each participant was given a blank external hard
drive, which stayed with them throughout the test, was dedicated to transfer data to the
repository at the end of each test. Blank DVD/CDs were also made available for those who
wanted to copy the data to DVDs, then the data were copied to the repository (through the
repository desktop).
Data was provided from each participant in National Maritime Electronics Association (NMEA)
0183, Receiver Independent Exchange (RINEX versions 2.11, 3.00, 3.01 and 3.03), or
commonly defined comma separated variable format (CSV). The preference was to have data
provided in CSV format when possible. Table 3-7 identifies the desired data from each GNSS
receiver tested. It was understood that all data types may not be available and for these instances
fields should be denoted not available, “NaN”. The CSV format accommodates twelve fields to
indicate GPS time, position estimate and satellites tracked for each GNSS constellation and
signal type. This data is followed with satellite specific data needed for the analysis. Each
constellation signal type is allocated 32 satellites and SBAS is allocated two satellites with the
data grouped by data block as described in Table 3-7.
Each receiver has a separate data file for each test run. The nomenclature of the file name is as
follows: ParticipantLC_ParticipantID_Test#_Date.extension, where ParticipantLC is a unique
indicator for the antenna location (and receiver if integrated), ParticipantID maps to the
receiver/antenna tested and origination, Test# indicates the test run number for that day, Date is
the day of the actual test, and an extension is used to indicate the type file (ex. NMEA, RINEX
or CSV). The ParticipantLC and ID were provided during test set-up. After the test week all the
data files were converted to a commonly defined CSV as well as MAT format, as shown in Table
3-7 using MATLAB. To facilitate post-data processing, MATLAB Datenum and GPS Week
columns were added at the beginning of the table, and the file’s naming convention was changed
to add device under test number (DUT#), participant’s acronym, receiver’s name, and the
category for receiver category.

44
Table 3-7: CSV Data Format
Field # Parameter Units Comment
1 GPS Seconds of Week Seconds
2 Latitude Degrees relative to WGS84
3 Longitude Degrees relative to WGS84
4 Height Meters relative to WGS84 (orthometric)
5 GPS L1 C/A-code Tracked
6 GPS L1 P-code Tracked
7 GPS L1C Tracked
8 GPS L2 P-code Tracked
9 GLONASS L1 C Tracked
10 GLONASS L1 P Tracked
11 BeiDou BI1 Tracked
12 Galileo E1 B/C Tracked
13 SBAS L1 C/A-code Tracked
Next are 32 satellite slots per signal type for GPS, GLONASS, BeiDou and Galileo. Each slot contains four data
parameters. Satellite order is 1 through 32. SBAS has two satellite slots with four data parameters each. SBAS
pseudo random noise (PRN) codes are 135 and 138.
14 Carrier Phase(PRN-1,GPS C/A) meters
15 Loss of Lock Flag(PRN-1,GPS C/A) binary (0 or 1) Cycle slip or loss of carrier phase lock
indicator. 0 indicates no loss of lock, 1
means lost lock.
16 Carrier to Noise Ratio(PRN-1,GPS
C/A)
dB-Hz
17 Pseudorange(PRN-1,GPS C/A) meters
18 – 141 GPS L1 C/A-code measurements for satellite signals 2-32
142 – 269 GPS L1 P-code measurements for satellite signals 1-32
270 – 397 GPS L1C measurements for satellite signals 1-32
398 – 525 GPS L2 P-code measurements for satellite signals 1-32
526 – 653 GLONASS L1 C satellite signals 1-32
654 – 781 GLONASS L1 P satellite signals 1-32
782 – 909 BeiDou BI1 satellite signals 1-32
910 – 1037 Galileo E1 B/C measurements for satellite signals 1-32
1038 - 1045 SBAS L1 C/A-code measurements for satellites signals from PRNs 135 and 138
1 dB CNR Degradation
The 1 dB carrier-to-noise ratio (CNR) interference protection criterion (IPC) has been used in
responding to FCC rulemaking proceedings that assessed the potential impact to GPS services,
[2] and was the subject of much discussion and stakeholder feedback at the OST-R workshops.

45
A 1 dB C/N0 degradation (-1 dB C/N0) due to a new interference source is equivalent to an I0/N0
ratio of -6 dB, where C is the level of the observable desired information signal, while N0 is the
competing unwanted noise and I0 is the interference level. This I0/N0 ratio of -6 dB means that a
new interference level is actually one fourth the level of the existing noise level and the total
unwanted N0 + I0 level is now 25% higher which is highly significant to system designers.
There are multiple interference mechanisms that can degrade C/N0 of a GPS receiver. However,
it is difficult to isolate the specific interference mechanism for each GPS receiver without
sufficient technical information, such as receiver design, radio frequency filter selectivity, low
noise amplifier gain, noise figure, 1 dB gain compression point and third-order intercept point
from the GPS receiver manufacturers. Participation by GPS/GNSS receiver manufacturers in the
DOT GPS Adjacent Band effort was on a voluntary basis and there was no obligation to provide
this information.
Given the myriad of GPS/GNSS applications requiring accuracy to support their mission
applications ranging from tens of meters to millimeters, there is not a single "accuracy
degradation limit" that could be applied and trying to do so would be an intractable effort.
ITM Data Processing
The Interference Tolerance Mask (ITM) defines, for a given frequency, the maximum aggregate
power allowed to ensure the tested GPS/GNSS receiver did not experience more than a 1 dB
reduction in CNR for various categories of GPS/GNSS receivers. For a given DUT, the
interference power (IP) data was calibrated using the mapping measurements interpolated to the
DUT location. The CNR data corresponding to the GNSS signal being analyzed is time aligned
with the calibrated interference data.

46
Figure 3-16: ITM Processing Block Diagram
The IP level which causes 1-dB CNR degradation is then determined on a per PRN basis. The
median of results across PRNs is the value of the interference tolerance mask 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) at
frequency f for that DUT. The use of mean and median produced similar results but the median
operation was chosen because it is less sensitive to outliers. Only the PRNs at minimum ICD
powers were used in the ITM analysis (not lower power to emulate low elevation satellites). A
description of the overall processing is outlined in Figure 3-16.
Figure 3-17 shows the time aligned IP and reduced CNR data. The baseline CNR (𝐶𝐶𝐶𝐶𝑅𝑅𝐵𝐵𝐵𝐵,
magenta dashed line) is calculated by averaging over the last 2.5 minutes of the IP-off interval
(black line). During the IP-on interval, data reduction was performed by averaging CNR over the
last 12 seconds of each 15 seconds IP step in order to allow for three seconds settling time. The
blue trace in this figure is the resulting reduced CNR time series.
𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) = 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀_𝑘𝑘 {𝐼𝐼𝐼𝐼𝐼𝐼_𝑘𝑘 (𝑓𝑓) }
(k = 1 to NSV)
Receiver Data
Receiver Data
Downselection
Calculate uncertainty on
mean 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) as
𝜎𝜎𝐼𝐼𝐼𝐼𝐼𝐼
�NSV
Interference
Power Data
Calibration
Data
Data Selection
Parameters
- Test Number
- GNSS Signal
Type
- Interference
Signal Type and
Analysis
Frequency
- Receiver
Location
IP Data Calibration
and Downselection
Time Alignment and
Data Reduction
𝐼𝐼𝐼𝐼𝑀𝑀𝑘𝑘(𝑓𝑓) = 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓, 𝑃𝑃𝑃𝑃𝑁𝑁𝑘𝑘)
Computation Module
(k = 1 to NSV)

47
Figure 3-17: Sample plot for calibrated interference power overlaid with time aligned
CNR data for a given DUT at a particular interference frequency
The time aligned and reduced CNR data can be plotted directly as a function of IP for each 𝑃𝑃𝑃𝑃𝑃𝑃
and 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓, 𝑃𝑃𝑃𝑃𝑁𝑁𝑘𝑘) can be found using linear interpolation as shown in Figure 3-18.
Figure 3-18: Determining the tolerable interference level from the CNR versus interference
power for a single satellite after time alignment and calibration of interference power
The test for each interference signal type was performed twice as described in Section 3.1.5. The
average of the resulting two interference tolerance values was taken as the final 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) for
each interference signal type and GNSS signal supported by a DUT. Figure 3-19 depicts the
ITMs for L1 C/A signal for a single DUT and shows repeatability between both LTE tests.
IP-Off IP-On
-70 -65 -60 -55 -50
Interference Power (dBm)
39
40
41
42
43
44
45
46
CNR
(dB-Hz)
CNR Measurement
CNR Interpolated
IP at Baseline CNR - CNR = 1dB
Baseline CNR
X: -61.66
Y: 44.18
1 (dB)
ITM (f,PRNk)

48
Figure 3-19: Overlaid L1 C/A ITMs from two radiated LTE test events for a single DUT.
Test-2 and Test-3 refer to the first and second LTE tests respectively.
It is important to note that the algorithm does not calculate a 1-dB CNR degradation value if the
CNR dynamic range defined as the difference between 𝐶𝐶𝑁𝑁𝑅𝑅𝐵𝐵𝐵𝐵 and the smallest 12 second
average CNR value within each IP progression (𝐶𝐶𝐶𝐶𝑅𝑅𝑀𝑀𝑀𝑀𝑀𝑀) is less than 1.5 dB. In addition, the
algorithm also checks that this dynamic range is statistically significant. This is done by ensuring
that the standard deviation of the difference is small relative to its mean value. This criterion
used by the algorithm is shown in the following expression (in dB).
𝑪𝑪𝑪𝑪𝑹𝑹𝑩𝑩𝑩𝑩 − 𝑪𝑪𝑪𝑪𝑹𝑹𝑴𝑴𝑴𝑴𝑴𝑴 > 𝟑𝟑 × ��𝒗𝒗𝒗𝒗𝒗𝒗(𝑪𝑪𝑪𝑪𝑹𝑹𝑩𝑩𝑩𝑩) + 𝒗𝒗𝒗𝒗𝒗𝒗(𝑪𝑪𝑪𝑪𝑹𝑹𝑴𝑴𝑴𝑴𝑴𝑴)� (3-1)
Where 𝑣𝑣𝑣𝑣𝑣𝑣(. ) represents the measurement variance divided by the number of measurements for
each of the two quantities. Additional quality checks are performed for each IP step. For
example, the algorithm requires that at least three measurements be reported within the last 12
seconds interval of each step and that the standard deviation of the mean within each step be less
than ½ dB.
The averaging across repeated tests was subject to additional quality control checks. For each
DUT, GNSS signal, interference type, and center frequency combination, this average was
performed only when the difference between the results produced by the two tests is less than 10
dB. For the cases when the difference exceeded this threshold, the test producing a value closer
to the interpolated value between adjacent frequencies is kept, and the result of the other test was
disregarded as an outlier.
The differences in results between repeated tests were analyzed as a measure of uncertainty due
to environmental and equipment variability and is a real measure of test repeatability. Another
uncertainty measure was calculated by considering the variability of 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓, 𝑃𝑃𝑃𝑃𝑁𝑁𝑘𝑘) across PRNs
for each DUT as shown. The empirical cumulative distribution function (CDF) for both error

49
quantities just described are shown in Figure 3-20 for the 10 MHz LTE and Figure 3-21 for the 1
MHz AWGN interference signals. These plots show that 90 percentile of the uncertainty is less
than 1.5 dB based on PRN variability analysis and less than 3 dB in terms of test repeatability.
This is a near upper bound estimate on the measurement error. The median of both uncertainty
measures are less than 0.5 dB.
Figure 3-20: CDF of measurement uncertainty calculate from per DUT differences across
PRNs (black) and test to test difference (red) for the 10 MHz LTE interference signal
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
X: 2.828
Y: 0.9
Difference, dB
Quantile,
fraction
Measurement Uncertainty
X: 1.128
Y: 0.9
PRN-to-PRN
Test-to-Test

50
Figure 3-21: CDF of measurement uncertainty calculate from per DUT differences across
PRNs (black) and test to test difference (red) for the 1 MHz AWGN interference signal
ITM Aggregation and Test Results
The bounding ITM mask is the one that protects all receivers within a category. The value of the
bounding ITM at each frequency is found by taking the minimum of ITM(f) across all receivers
in the category (i.e. at a particular frequency, this is the smallest interference power that causes
1-dB degradation for any receiver in the category). Multiple bounding ITMs are generated based
by determining one for each interference signal type and GNSS signal combination. The
bounding masks for each category corresponding to the 10 MHz LTE interference signal and L1
C/A GPS signal are shown in Figure 3-22. This plot shows the HPR and SPB categories to be the
most susceptible in terms of received interference power levels with the cellular category
generally being the most tolerant of LTE interference.
10
-4
10
-3
10
-2
10
-1
10
0
10
1
10
2
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
X: 2.828
Y: 0.9
Difference, dB
Quantile,
fraction
Measurement Uncertainty
X: 1.372
Y: 0.9
PRN-to-PRN
Test-to-Test

51
Figure 3-22: GPS L1 C/A bounding ITM for each category of receivers
These ITMs are also calculated for all other emulated GNSS signals. This is shown in Figure
3-23 for the HPR receiver category. Interference power levels from a 10 MHz LTE type signal
should not exceed any of the masks depicted in this figure if all GNSS signals are to be protected
for the HPR category.

52
Figure 3-23: HPR bounding ITMs for each of the emulated GNSS signals
Figure 3-24 overlays the HPR bounding ITMs corresponding to both the 10 MHz LTE
interference signal (solid lines) and the 1MHz AWGN interference signal (dotted lines). In
general, the results show a weak dependence for the bounding ITMs on interference signal type
helping to further generalize the results beyond the LTE type signal if needed. An exception to
that is the GLONASS L1 P bounding ITM that shows up to 10 dB more sensitivity to the 10
MHz LTE signal. This is likely due to one or more receivers processing GLONASS L1 P signal
that did not collect valid data during the 1 MHz interference signal test.
1450 1500 1550 1600 1650 1700
Frq(MHz)
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
IP
at
1dB
(dBm) 10MHz-GPS L1CA
10MHz-GPS L1P
10MHz-SBAS L1CA
10MHz-GLONASS L1C
10MHz-GLONASS L1P
10MHz-BeiDou B1I
10MHz-Galileo E1BC

53
Figure 3-24: HPR Bounding ITMs for each of the emulated GNSS signals. ITM bounding
masks for the 1 MHz AWGN and 10 MHz LTE interference signals are shown
Figure 3-25 (a) shows the aggregate result for the HPR category and L1 C/A GPS signal type.
The lower and upper bounds, as well as the various percentile levels are presented to give an
indication of the data distribution. The lower the percentile level the more protection it offers.
For example, the 10th percentile indicates the received interference power level that leaves 10%
of the tested receivers unprotected while the 90th percentile is the value that leaves 90% of the
receivers unprotected. In order to ensure tolerable level of interference to all tested receivers only
the lower bound (or minimum value) is considered. Figure 3-25 (b) shows the same percentile
results but for ITMs that protect all emulated GNSS signals processed by the tested HPR
receivers. This is done by first calculating the minimum ITM across the supported GNSS signals
for each DUT and then calculating the various percentiles across DUTs. These two plots indicate
that the interference power levels needed to protect all GNSS signals are generally lower but
comparable the L1 C/A ITM levels for the tested receivers.
1450 1500 1550 1600 1650 1700
Frq(MHz)
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
IP
at
1dB
(dBm) 1MHz-GPS L1CA
1MHz-GPS L1P
1MHz-SBAS L1CA
1MHz-GLONASS L1C
1MHz-GLONASS L1P
1MHz-BeiDou B1I
1MHz-Galileo E1BC
10MHz-GPS L1CA
10MHz-GPS L1P
10MHz-SBAS L1CA
10MHz-GLONASS L1C
10MHz-GLONASS L1P
10MHz-BeiDou B1I
10MHz-Galileo E1BC

54
(a)
(b)
Figure 3-25: 10 MHz Statistical Mask Results for
High Precision receivers for (a) GPS L1 C/A (b) All Emulated GNSS Signals.

55
A comprehensive set of bounding and statistical ITMs have been produced for all receiver
categories and GNSS signal type combinations, and are shown grouped by interference signal
type in Appendix B.
These bounding masks can then be used in an inverse modeling analysis to compute the tolerable
transmitter EIRP levels corresponding to a given transmit application and use-case parameters.
In particular, the bounding masks for the L1 C/A GPS signals are used later in this report to
calculate tolerable EIRP levels by receiver application.
Loss of Lock Data Processing
As discussed in Section 3.1.5.2, the ITMs are the interference levels that resulted in a 1 dB
degradation in CNR. As illustrated in Figure 3-26, most receivers continued to report CNR
measurements after the interference level exceeded the ITM. Within this report, “loss of lock” is
defined to be situation in which the interference increased to the point where the receiver ceased
reporting CNR for a particular signal and a particular satellite. The “loss of lock” point is
interpreted herein to mean that the DUT is no longer able to track that signal type (i.e. L1 C/A).
Figure 3-26: Determination of Loss of Lock Interference Level from CNR Data
The processing of CNR to yield interference levels corresponding to loss-of-lock was consistent
with the processing used to determine ITMs, with the exception illustrated in Figure 3-26.
Namely, that the loss-of-lock interference level was determined based upon the highest level of
interference for each signal/satellite for which the DUT reported a CNR value. Although, as
-70 -65 -60 -55 -50
Interference Power (dBm)
39
40
41
42
43
44
45
46
CNR
(dB-Hz)
CNR Measurement
CNR Interpolated
IP at Baseline CNR - CNR = 1dB
Baseline CNR
IP at loss of lock

56
discussed in Section 3.1.6.1, collected data for each DUT included a “loss-of-lock indicator” this
data was found to be unreliable, not available, or inconsistent amongst DUTs. Therefore, the
approach outlined above was adopted to determine the loss-of-lock interference levels.
Two loss-of-lock levels were determined for each DUT, for each interference type, and for each
interference frequency:
1. High-elevation satellite – this interference level corresponded to loss-of-lock for the
nominally powered GNSS signals (See Table 3-4), i.e., the ones that were not attenuated
by 10 dB or 20 dB with respect to the specified minimums in applicable Interface Control
Documents or Interface Specifications. This interference level was averaged across all
applicable (up to 10) satellites.
2. Low-elevation satellite – this interference level corresponded to loss-of-lock for the
GNSS signals that were 10 dB below nominal (see Table 3-4). Only one such signal was
broadcast for each GNSS constellation (GPS, GLONASS, Galileo, BeiDou). The “low-
elevation” designation is appropriate, since as discussed in Section 3.3 typical DUT
antennas provide approximately 10 dB less gain towards low-elevation angle satellites
than they do towards zenith. In the chamber testing, the GNSS transmit antenna was at
zenith so all GNSS signals arrived from zenith in the testing. This situation is different
from the real-world, in which GNSS signals can arrive from all elevation angles in the
upper hemisphere.
The loss-of-lock levels computed using the above method should be viewed as the received
interference power levels for which there is very high confidence that high- or low-elevation
angle satellites are completely unusable by a GPS/GNSS receiver. These estimated loss-of-lock
levels may be significant overbounds for several reasons including:
• As noted in Section 3.1.4.1, DUTs in the center of the test grid experienced received
GNSS signal levels that were more than 3 dB greater than the minimum specified levels
for each GNSS signal type. If they were presented with minimum specified GNSS signal
levels, it is likely that these DUTs would lose lock on GNSS signals in the presence of
lower levels of interference.
• It is likely that many DUTs continued to track and output C/N0 for satellites that would
no longer be useful for navigation due to poor tracking quality. For instance, many DUTs
reported GPS C/A-code C/N0’s below 20 dB-Hz. The GPS C/A-code signal includes 50
bps navigation data that is unencoded (i.e., no forward error correction is utilized). At 20
dB-Hz, the bit-energy to noise density Eb/N0 is 3 dB and it is not theoretically possible to
read the navigation data as necessary for positioning without external assistance. With an
Eb/N0 of 3 dB, it is unlikely that the DUT could track carrier phase to provide a coherent
phase reference, but even if it could the probability of correctly decoding a 300-bit GPS
navigation data subframe without error is less than 0.0001.
Figure 3-27 and Figure 3-28 show interference powers resulting in loss-of-lock for high-
elevation and low-elevation angle satellite GPS C/A-code signals, respectively. The interference
powers resulting in loss-of-lock for high elevation angle satellites were typically 15 – 25 dB

57
higher than 1 dB ITMs. The interference powers resulting in loss-of-lock for low elevation angle
satellites were typically 5 – 15 dB higher than 1 dB ITMs.
Figure 3-27: Interference Power resulting in Loss of Lock for GPS L1 C/A-code (High
Elevation Angle).

58
Figure 3-28: Interference Power resulting in Loss of Lock
for GPS L1 C/A-code (Low Elevation Angle).
Additional loss of lock results are provided in Appendix C.
3.2 Conducted (Wired) Testing
Wired tests were executed subsequent to WSMR radiated tests for specialized scenarios suited to
a laboratory environment. This testing was conducted during July 2016 at Zeta Associates Inc.
in Fairfax, VA. The specific objectives for wired testing included: (1) evaluation of the impact
of adjacent-band interference on signal acquisition, (2) comparison between wired and radiated
receiver susceptibility to adjacent band interference with 1 MHz bandpass noise and 10 MHz
LTE (same signals as used in the anechoic chamber), and (3) assessment of the impact of an
adjacent band transmitter noise floor (out-of-band to the interference source, in-band to
GPS/GNSS) in addition to the fundamental emission.
Devices Test
For this testing, fourteen of the 80 receivers tested at WSMR were selected and provided by
USG partners. These receivers covered all GPS/GNSS categories from WSMR except Space
Based. The receiver categorization, and specific port location are given in Table 3-8. Notice the
majority of receivers tested were from the high precision category.

59
Table 3-8: Receivers Tested
8-Way Splitter w/Individual Amps. After 8-Way Splitter w/Single Amp. in Front
1. Monitor (spectrum analyzer) 1. TIM
2. HPR 2. GAV
3. HPR 3. HPR
4. CEL 4. HPR
5. GLN 5. HPR
6. HPR 6. HPR
7. HPR 7. HPR
8. Monitor (spiral enclosure) 8. HPR
Signal Generation
Wired testing utilized the same core signal generation equipment as used for radiated testing at
WSMR. The conducted circuit is shown in Figure 3-29. In addition to the GNSS playback and
SPIGAT systems used at WSMR, a circuit was added for simulating out-of-band emissions
(OOBE--lower left in figure where this is added in-band noise to GNSS but out-of-band for
fundamental interference signal). The interference, GNSS and OOBE signals are added by a
power combiner and conducted to the devices under test via multi-port power splitters with an
isolator at each port to prevent port interaction. After the isolator, a broadband LNA provides
necessary gain as a substitute for the active antenna in the radiated environment, with test power
referenced to the LNA input. To allow static configuration throughout testing, adjustable
attenuators were included on the GNSS and OOBE signal paths not only to set proper levels but
also to serve as switches (when at high attenuation) for complete removal of these signals as
necessary. Other modifications to the interference system included the substitution of a lower
power HPA (more than adequate for the reduced attenuation of the conducted path) and a notch
filter targeting the RNSS band.

60
Figure 3-29: Interference System Configuration for Wired Tests
Signal Acquisition
Signal acquisition tests were executed at four adjacent-band frequencies using LTE signals at
1525, 1550, 1620 and 1645 MHz. The test sequences removed the GNSS signals for 30 seconds
and then allowed at least 90 seconds after they were reintroduced for the receiver to reacquire
and track. (The original test plan used 120 seconds to allow GNSS signals to be reacquired, but
after analyzing pre-test data it was determined this time could be shortened to 90 seconds to
expedite the test.) These tests are therefore more indicative of Warm or Hot Start versus the
potentially more challenging acquisition condition of Cold Start. This sequence of removing and
reintroducing signals was repeated in sets of five iterations starting with a set where interference
was turned OFF. After this quiescent period, the interference was turned ON and after each
successive completion of five iterations its power was incremented by 2 dB. Interference power
ranged from -60 to -10 dBm for the outer two frequencies (1525 and 1645 MHz) and -80 to -30
dBm for the inner two frequencies (1550 and 1620 MHz). The maximum power tested in each
range matched the maximum power used in the baseline LTE tests for these frequencies.
Out-of-Band Emissions
Out-Of-Band Emissions (OOBE) refer to the emissions from adjacent frequency band terrestrial
deployments into the 1559-1610 MHz band. For OOBE tests, the OOBE circuit generated a flat
wideband noise pedestal centered on the RNSS band with spectral density controlled by the
programmable attenuator. OOBE density levels used for testing were defined by associating the
LTE power levels at the specified maximums of 62 dBm (32 dBW) for base stations and 23 dBm
(-7 dBW) for handsets with each wideband OOBE limit as summarized in Table 3-9. The
LTE/OOBE ratio is defined at these limits and applied (added) to the target LTE signal power at
each point in the test to determine the corresponding OOBE level that should be received.
Figure 3-30 depicts the relationship between the OOBE (in dBW/Hz) and LTE power levels (in
dBm) at the receiver’s RF input port. This relationship is linear with a slope of one since OOBE

61
and LTE signal powers undergo the same path loss (neglecting the slight dependence of path loss
on frequency). This figure also shows the approximate OOBE level (horizontal dashed line) for a
receiver noise floor of -201.5 dBW/Hz and associated LTE receive power levels (intersection of
the dashed horizontal line and LTE receive power vs. received OOBE lines) which would cause
a 1 dB CNR degradation for the various handset and base station limits outlined in Table 3-9.
Additional details on the conducted testing OOBE levels are provided in Appendix D.

62
Table 3-9: Ratio of OOBE limit density to
LTE power for setting OOBE testing levels
OOBE density LTE power
ratio
OOBE/LTE
[dBW/MHz] [dBW] [dB/MHz]
FCC base station* -70 32 -102
FCC handset* -70 -7 -63
Proposed base
station** -100 32 -132
Proposed handset** -105 -7 -98
* Based upon FCC Mobile Satellite Service Ancillary Terrestrial Component (ATC)
rules, contained within Title 47 of the Code of Federal Regulations, Part 25.
** Based upon characteristics of a proposed adjacent-band LTE network.
Figure 3-30: OOBE Levels Associated with LTE Signal Power used in Testing
System Calibration
Calibration of SPIGAT was accomplished in the same fashion as described for WSMR with 22
CW tones at each frequency and measuring with a spectrum analyzer to generate a bias table.
The spectrum analyzer in this instance was connected to the power splitter versus the cavity
backed spiral in the chamber. The GNSS playback signal levels were verified by showing CNR
estimates from the GPS receiver used for monitoring at WSMR matched the levels observed in
this laboratory setting.

63
Test Sequence
Wired tests were executed for baseline 1 MHz bandpass noise and 10 MHz LTE signals, FCC and
proposed OOBE levels, and signal acquisition. Tests were numbered 10 through 18 with Table
3-10 summarizing the test schedule.
Table 3-10: Wired Test and Data Summary
Day of Week (25-29 July 2016) Test and Number
Monday Type 2 – Test10
Type 1 – Test11
Tuesday Type 2 w/OOBE FCC – Test12
Acq. @ 1525 MHz – Test13 (120 sec. dwell)
Wednesday Test14 stopped early/network issue
Acq. @ 1620 MHz – Test15
Thursday Acq. @ 1645 MHz – Test16
Acq. @ 1550 MHz – Test17
Friday Type 2 w/Proposed OOBE – Test18
Data Processing
The following sections detail results from the wired testing for GPS L1 C/A processing only.
Processing for wired results followed the approach discussed above for determining 1 dB CNR
degradations as a function of interference power. Signal acquisition processing required its own
considerations and is discussed in that section.
Comparison Tests
The comparison tests were intended to demonstrate equivalence between the radiated and wired
test environments. Two example results of interference power causing 1 dB CNR degradation
are shown in Figure 3-31 for the LTE interference signal. Example (a) in this figure compares
results for a high precision receiver. Here performance matches very well for frequencies closest
to the RNSS band while for frequencies further away the radiated performance is superior. This
divergence is an expected result since at WSMR the receiver used its native antenna which
included some filtering (along with a low noise amplifier) which served to suppress peripheral
interference. The difference, therefore, is directly related to not having this filter/LNA module
available in the wired testing. Example (b) of that same figure is a case where the antenna was
integrated with the enclosure. For wired testing the signal could be inserted directly after the
passive element. In this instance, the radiated and wired results match very closely because both
include all components influencing mitigation of adjacent interference. In general, comparisons
of radiated and wired tests showed expected agreement with differences attributable to bypassing
of active antennas in the wired test.

64
(a) (b)
Figure 3-31: Comparison of IP causing 1 dB degradation for the LTE Interference Signal
from Radiated and Wired Testing for: (a) High Precision receiver and (b) Cellular device.
OOBE Results
Tests conducted with OOBE were executed by adding noise in the RNSS band as shown in the
wired test description. Figure 3-32 provides examples of two receivers with significantly
different rejection performance for adjacent band interference. These examples show baseline
(wired) performance of interference power causing a 1 dB degradation contrasted with OOBE
performance for FCC prescribed and proposed levels for one applicant. As Figure 3-30
predicted, receiver performance can be impacted by inclusion of OOBE at FCC base station and
FCC handset levels. This result is clearly demonstrated in (b) since this receiver provides good
rejection of adjacent interference and therefore inclusion of noise in the RNSS band results in 1
dB CNR degradation not observed with the baseline test. The most distinct difference in
performance is evident for handset frequencies, where adding OOBE at the FCC limits result in 1
dB CNR degradation at approximately -50 dBm compared with much more robust performance
when OOBE is not included. The proposed base station OOBE limits did not result in 1 dB CNR
degradation with the LTE power levels tested and for proposed handset limits the 1 dB CNR
degradation level was observed at approximately -15 dBm : In context of distance and presuming
complete rejection of adjacent band interference, the proposed OOBE limits for base station and
handsets suggest 1 dB CNR degradation could be expected within approximately 4 meters (3.5
m) and 2 meters, respectively. These numbers were obtained for a receiver noise level of -201.5
dBW/Hz and assuming free space path loss and an omnidirectional transmitter antenna gain
pattern.

65
Figure 3-32: Interference power causing 1 dB CNR degradation for baseline and OOBE
tests for (a) High Precision receiver and (b) Cellular device.
Acquisition Results
Receiver acquisition tests were processed to show both average acquisition time, and the
interference power level when receivers could no longer acquire. Acquisition time was
computed for L1 C/A signals at the specified minimum power level (-128.5 dBm for L1 C/A)
and also for one satellite that was set 10 dB lower to represent low elevation or challenged
environments. For the specified minimum signals, the acquisition time was defined as the
receiver acquiring and tracking four or more of these satellites. Since more than four satellites
are generally in view at the specified minimum levels this is considered a modest criterion for
establishing acquisition. For the low elevation satellite, the acquisition time was simply when
this satellite was first acquired and tracked. At each interference power level, acquisition time
from the five iterations was averaged to provide a single value. Note that at each power step an
acquisition time was computed only if the receiver met the acquisition criterion for all five
iterations.
The results from all receivers tested were compiled for each test frequency and are shown in
Figure 3-33 through Figure 3-36. Figure “(a)” provide the number of receivers satisfying the
acquisition criteria for specified minimum (“ICD Min. Power”) and low elevation (“Low
Elevation”) signals. Additionally, the figure shows the number of receivers at each interference
power step where the CNR degradation is less than 1 dB (“IP @ 1 dB”). For example in Figure
3-33 (a), there were two receiver where the interference power associated with their 1 dB CNR
degradation was less than the starting power level of -60 dBm and therefore this count starts at
ten. This figure shows further that this count of “IP at 1 dB” appears closely associated with the
number of receivers capable of acquiring the signal emulating low elevation conditions.
(a) (b)

66
For the GPS receivers tested, the 1 dB C/N0 degradation point can be an indicator of negative
impact to signal acquisition time in low elevation satellite conditions.. Figure “(b)” show average
acquisition time for specified minimum and low elevation signals and generally demonstrate
acquisition degradation with increasing interference power.
Figure 3-33: Summary acquisition performance for 1525 MHz for ICD minimum and low-
elevation satellites. (a) Number of DUTs, and (b) average acquisition time.
elevation satellites. (a) Number of DUTs and (b) average acquisition time.
(a) (b)
(a) (b)

67
3.3 Antenna Characterization
Twenty GNSS antennas, most of which were involved in the WSMR testing, were characterized
with respect to frequency selectivity, elevation pattern, and RF gain/compression characteristics
through anechoic chamber/live-sky/RF measurements in order to help interpret the WSMR test
results and facilitate the calculation of tolerable transmit power.
A representative set of antennas was characterized though a set of activities including:
1) Anechoic chamber measurements - From June through August 2016, the gain patterns for
14 external antennas were measured in an anechoic chamber at MITRE in Bedford, MA.
(a) (b)
(a) (b)

68
2) Live-sky C/N0 measurements – In August 2016 and February 2017, the relative gain
patterns of five antennas that were integrated with GNSS receivers were estimated using
live-sky GPS C/A-code C/N0 measurements.
3) Active sub-assembly measurements - From August through October 2016, the gain and
compression characteristics of the active subassemblies of four external antennas were
measured at Zeta Associates in Fairfax, VA.
The following subsections describes these antenna characterization activities and the resultant
measurements.
Selected Antennas
Table 3-11 lists the antennas that were characterized.
Table 3-11: Characterized GNSS Antennas
Manufacturer Model Characterization Approach
AeroAntenna AT575-142-614-50 Anechoic chamber
AeroAntenna AT2775-42SYW Anechoic chamber
Arbiter AS0087800 Anechoic chamber
Garmin EDGE 1000 Live-sky C/N0 measurement
Garmin eTrex 20x Live-sky C/N0 measurement
Garmin GA-25 Anechoic chamber
Garmin GA-38 Anechoic chamber
Garmin GPSMAP 64 Live-sky C/N0 measurement
Hemisphere 804-3059-0 Anechoic chamber
Javad Triumph-1 Live-sky C/N0 measurement
Leica AX1202GG Anechoic chamber and active sub-
assembly measurements
Navcom 82-001020-3001LF Anechoic chamber
PCTel 3977D Anechoic chamber
Samsung S5 Live-sky C/N0 measurement
Trimble Bullet 360 Antenna
101155-10
Anechoic chamber
Trimble Choke Ring 29659-00 Anechoic chamber and active sub-
Trimble Zephyr 41249-00 Anechoic chamber
Trimble Zephyr Geodetic 2 55971-
00
Anechoic chamber and active sub-
Trimble TRM59800 module Active sub-assembly measurements
u-blox ANN-MS-0-005 Anechoic chamber
Chamber Measurements
Two-dimensional (elevation and azimuth) gain patterns for incident signals of four polarization
types, right hand circularly polarized (RHCP), left hand circularly polarized (LHCP), vertically
polarized (V), and horizontally polarized (H) were measured at 22 frequencies: 1475, 1490,
1495, 1505, 1520, 1530, 1535, 1540, 1545, 1550, 1555, 1575, 1595, 1615, 1620, 1625, 1630,
1635, 1640, 1645, 1660, and 1675 MHz. The measurements were made in a 30 ft × 21 ft × 15 ft

69
anechoic chamber at MITRE in Bedford, MA. A calibrated, automated antenna measurement
system developed by Nearfield Systems was utilized.
All 14 antennas were active, and the gains measured were thus a combination of passive element
gain and amplifier gain. Absolute gain of the passive elements of each active antenna was not
directly observable without breaking into the antennas. Antenna directivity, however, was
calculated from the patterns using Nearfield Systems’ NSI2000 software.
Figure 3-37 and Figure 3-38 provide some example results. Figure 3-37 shows the frequency
selectivity of the 14 antennas for incident RHCP signals as seen at antenna boresight. Note the
wide variation in selectivity to adjacent-band signals. Figure 3-38 shows the relative RHCP
antenna gain vs off-boresight angle at 1575.42 MHz for the 14 antennas. Each curve is
normalized to 0 dBic gain at boresight. In this figure, each point in the plotted results represents
an average across 180 deg of azimuth. The red vertical lines correspond to 5 deg elevation angle
on either side of the antenna. Additional measurements are provided in Appendix E.
The boresight directivities of the 14 measured antennas for RHCP signals at 1575.42 MHz varied
from 3.2 dBic to 8.0 dBic with a mean directivity of 5.4 dBic. Assuming 90% efficiency for all
of the antennas yields a rough estimate for passive element gains ranging from 2.7 dBic to 7.5
dBic with a mean of 4.7 dBic. In the WSMR radiated chamber testing, all of the tested receivers’
antennas were boresighted at the transmitting GNSS signal and interference generator antennas.
The measured relative antenna gain patterns can be utilized to model what gains would be seen
towards GNSS satellites and interference sources at other elevation angles in the “real world”.
For instance, the results in Figure 3-38 justify the interpretation of the GNSS signal levels that
were generated at -10 dB power relative to specified minimum levels in the WSMR radiated
testing as corresponding to what would be seen in the real world towards low elevation angle
satellites for many of the tested antennas (note that the data in Figure 3-38 indicates relative
gains ranging from -3 dB to -15 dB for gain towards a satellite at 5 deg elevation above the
horizon vs gain towards a satellite at zenith).

70
Figure 3-37: Frequency Selectivity of the 14 External Antennas
Figure 3-38: Relative RHCP Gain Patterns of the 14 Antennas at 1575 MHz
(red vertical lines correspond to 5 deg elevation angle)
1450 1500 1550 1600 1650 1700
-120
-100
-80
-60
-40
-20
0
Frequency (MHz)
Relative
Attenuation
(dB)
-150 -100 -50 0 50 100 150
-40
-35
-30
-25
-20
-15
-10
-5
0
Off-boresight angle (deg)
Relative
Gain
(dBic)

71
Live-Sky Measurements
Some of the GNSS receivers tested at WSMR utilize integrated antennas. Estimates of their
relative antenna gain patterns at 1575 MHz were obtained through measurements of GPS C/A-
code C/N0 over short time intervals in an outdoor environment in two locations (rooftop of a
building at MITRE’s Bedford, MA complex and at Zeta’s Fairfax, VA location) with clear sky
views. Estimated relative gain pattern results using this method for four integrated GLN antennas
are shown in Figure 3-39 and for an integrated HPR antenna in Figure 3-40. Measured GPS C/A
C/N0 from a cellular device is shown in Figure 3-41 as a function of azimuth and elevation.
Based on analysis of this data and cellular GPS antenna design, placement and performance, the
use case analysis that follows simply assumed 0 dBi for antenna gain in all directions.
Figure 3-39: Relative L1 RHCP Antenna Gain Estimated from Live-sky C/N0
Measurements for Three GLN Integrated Antennas and Quadratic Fit
-80 -60 -40 -20 0 20 40 60 80
Off-boresight Angle (deg)
-25
-20
-15
-10
-5
0
Relative
Gain
(dBic)
Measurements
quadratic

72
Figure 3-40: Relative L1 RHCP Antenna Gain Estimated from Live-sky C/N0
Measurements for an Integrated HPR Antenna and Quadratic Fit
Figure 3-41: Live-sky C/N0 Measurements for a CEL Device
-80 -60 -40 -20 0 20 40 60 80
-20
-18
-16
-14
-12
-10
-8
-6
-4
-2
0
Relative
Gain
(dBic)
Measurements
Fit

73
Bench Test Measurements (Active Sub-assembly Measurements)
Bench test measurements were conducted on a further reduced set of antennas at Zeta to
characterize key filter/LNA performance parameters. The antennas were dissembled to access
the passive element connection to the filter/LNA assembly and then the response measured with
a network analyzer from approximately 1 GHz to 2 GHz. The LNA was powered for this testing
by inserting a bias-T on the RF output path and applying the required DC power. Results from
three antennas tested are shown in Figure 3-42 where each has been normalized for the measured
gain at GPS L1. These three antennas types were utilized at WSMR and as MITRE testing
demonstrated have vastly different characteristics presumably to meet their respective functions.
One of the filter/LNA devices was relatively narrowband and only passes GPS L1, another was
wider and clearly intended to pass both GPS and GLONASS L1, and lastly, the third device was
much wider and intended to pass MSS signals, GPS and GLONASS L1.
Figure 3-42: Normalized Filter/LNA Responses measured with Bench Testing
In addition to characterizing each assemblies response versus frequency, the devices were also
tested with a spectrum analyzer using “Intermod (TOI)” instrument software at the GPS L1
frequency to understand typical LNA characteristics of gain, 1dB compression (P1dB), input and
output third-order intercept points (IIP3 and OIP3) (in-band only). These results are shown in
Table 3-12 and again demonstrate the significant diversity observed with fielded GNSS
antennas.

74
Table 3-12: LNA Performance Characteristics measured with Bench Testing
Measurement at L1 Assembly #1 Assembly #2 Assembly #3
Gain (dB) 28 40 49
IIP3 (dBm) -12.7 -39.6 -29.3
OIP3 (dBm) 14.4 -1.4 17.7
Input P1dB (dBm) -25 -50 -42
The results of the antenna characterization indicate there is a very wide range of up to 80 dB in
selectivity farther away from the GPS band (1500/1650 MHz), which can explain the observed
50-dB range in IP for 1 dB CNR degradation. This variation in selectivity is most pronounced for
HPR devices, due to the fact that many HPR devices are designed to receive both GNSS signals
in the 1559 – 1610 MHz band as well as augmentation data over MSS satellites in the 1525 –
1559 MHz band. HPR devices that are not designed to process MSS signals also tend to utilize
wider bandwidths relative to other DUT categories to provide increased measurement precision
in the presence of multipath. There is a much narrower range up to 20 dB in selectivity close to
the GPS band (1550/1600 MHz) which is lower than the IP for 1 dB CNR degradation observed
for these frequencies.
Although changing antennas was not the focus of this effort, cost, viability, etc. should be
considered to determine the feasibility of such a solution. Antenna filtering can cause deleterious
effects on receiver performance, such as group delay and other distortions. Also, many HPR
DUTs are designed to receive augmentation data via MSS and to retain this functionality the
receiver passband needs to continue to extend into the applicable portion of the 1525 – 1559
MHz band.

75
4. TRANSMIT POWER LEVEL ANALYSIS (EXCLUDING CERTIFIED
AVIATION)
Approach
This section derives the transmit EIRP levels that can be tolerated by each category of
GPS/GNSS receivers except for certified aviation. Two complementary analyses are performed.
The first is a forward modeling approach that calculates the receive power map for a given EIRP
level and network deployment type. The receive power map is compared with the ITM(f), where
f is the frequency of interest, to identify the region where the corresponding category of receivers
is not protected from adjacent band interference. The second is an inverse modeling analysis that
calculates the tolerable EIRP for any given separation distance between the transmitter and
user’s receiver over a range of receiver heights.
For the transmit power level analysis, it is necessary to characterize the proposed transmitter
network deployment and the GPS/GNSS receiver use case scenarios. To ensure compatibility
with all receivers within a category, the bounding ITMs from Section 3 are used. The use of the
bounding ITMs ensures that the resultant EIRP values will protect 100% of the receivers tested.
Network Transmitter Parameters
Representative parameters for adjacent-band LTE networks were identified primarily from three
sources:
1) International Telecommunication Union Radiocommunication Sector (ITU-R) reports
and recommendations.
2) Federal Communication Commission (FCC) Mobile Satellite Service Ancillary
Terrestrial Component (ATC) rules, contained within Title 47 of the Code of Federal
Regulations, Part 25 (47 CFR 25).
3) Proposals for adjacent-band networks contained within FCC filings.
Base Stations
Report ITU-R M.2292 (henceforth “M.2292”) provides “Characteristics of terrestrial IMT-
Advanced systems for frequency sharing/interference analyses.” [3] For the 1-2 GHz band, these
include the characteristics listed in Table 4-1 for five deployment types.

76
Table 4-1: Base Station Characteristics from M.2292
Macro
rural
Macro
suburban
Macro
urban
Small cell
outdoor/Micro
urban
Small cell
indoor/Indoor
urban
Cell radius/
Deployment
density
> 3 km
(5 km
typical)
0.5 – 3 km
(1 km
typical)
0.25 – 1 km
(0.5 km
typical)
1 – 3 per urban
macro cell, <1
per suburban
macro site
Depending on indoor
coverage/capacity
demand
Antenna height 30 m 30 m 25 m 6 m 3 m
Sectorization 3 sectors 3 sectors 3 sectors Single sector Single sector
Downtilt 3 deg 6 deg 10 deg not applicable not applicable
Antenna
polarization
+/-45 deg +/-45 deg +/-45 deg Linear Linear
Maximum
antenna gain
18 dBi 16 dBi 16 dBi 5 dBi 0 dBi
Maximum
EIRP/sector*
61 dBm 59 dBm 59 dBm 40 dBm 24 dBm
*Values shown for this parameter are for 10-MHz LTE signals.
For macrocells, the cell radius and intersite distance are defined in Figure 4-1. Note that for
macrocells, the parameters A and B in the figure have the relationship B = 3A/2. For small cells,
each base station is located at the center of a cell resulting in an intersite distance 𝐵𝐵 = √3𝐴𝐴.
Figure 4-1: Macrocell Radius and Intersite Distance are A and B, respectively. Each
hexagon is referred to as either a sector or cell.
M.2292 references Recommendation ITU-R F.1336 (henceforth “F.1336”) for recommended
antenna gain patterns for each deployment type and provides input parameters. The normalized
gain patterns for the macro deployments are shown in Figure 4-2 and for the small cell
deployments in Figure 4-3.

77
Figure 4-2: Macro Base Station Antenna Gain Patterns (top – elevation; bottom – azimuth)
Figure 4-3: Small cell Base Station Antenna Gain Patterns (elevation patterns shown; both
patterns are omnidirectional in azimuth)
M.2292 indicates that base stations are only active 50% of the time, resulting in a time-average
EIRP that is half of the maximum value shown in Table 4-1. This EIRP reduction was not
utilized to be conservative, noting that a base station could be operating with 100% loading for
Small cell outdoor/Microurban Small cell indoor/Indoor urban
0
30
60
90
120
150
180
210
240
270
300
330
-10
-5
0
5
0
30
60
90
120
150
180
210
240
270
300
330
-10
-8
-6
-4
-2
0

78
long enough periods to cause disruptions to GNSS receivers. Also for conservatism, the “peak”
side-lobe gain patterns from F.1336 were used, as opposed to the “average” side-lobe gain
patterns. As suggested in F.1336, the average side-lobe patterns may be more appropriate for
studies involving an aggregation of base stations.
For the certified aviation analysis conducted by the FAA, a base station antenna gain pattern based
upon a specific LTE network proposal was utilized. See Appendix G.
Handsets
For handsets, M.2292 recommends modeling the antenna gain pattern as -3 dBi in all directions
with a maximum power supplied to the antenna of 23 dBm. This results in a handset model that
uses an isotropic antenna gain pattern with maximum EIRP of 20 dBm. Within this report, two
other EIRPs (still paired with an isotropic antenna assumption) are examined: 23 dBm and 30
dBm.
M.2292 indicates that handsets are active 50, 70, or 100% of the time depending on deployment
type. As with the base station models in this report, 100% handset activity was assumed for
conservatism since this level of activity can occur for short periods for any deployment type.
Further, M.2292 notes other factors that can diminish interference effects from handsets
including: power control that diminishes typical EIRPs by 21 – 32 dB, building shielding (up to
20 dB), and body shielding (4 dB). These three loss factors are also not considered within this
report for conservatism, since there are situations where none of the three losses may apply. For
instance, a handset can be outdoors at the edge of cell coverage transmitting maximum EIRP
towards a GPS/GNSS receiver without any intervening obstructions.
Use Case Development
Understanding GPS/GNSS receiver use cases scenarios are important so that the geometric
parameters, specifically a receiver height and lateral offset from a transmitter can be determined.
Also, it is important that use cases representative each receiver category and can provide a worst-
case scenario so most, if not all, receivers in that category are protected. In addition, use cases
are needed in conjunction with ITMs, propagation models, and transmitter scenarios to determine
what power levels can be tolerated adjacent to GPS/GNSS signals.
Use cases were compiled with input from DOT federal partners and agencies. Members of the
working group were provided a template that contained questions related to how their
organizations use GPS/GNSS receivers to support their mission. In particular, questions
included identifying height, speed, terrain, antenna orientation and integration, and urbanization
areas.
Also, outreach was conducted with GPS/GNSS receiver manufacturers. Manufacturers were
provided the same template as DOT’s federal partners. Additionally, manufacturers provided
presentations during several of the workshops that summarized use cases by category for the
receivers they manufacture.

79
A summary of the compiled results can be found in Table 4-2. The results generally indicate that
each category has a large range of geometric parameters.
Table 4-2: Summary of Compiled Use Case Information
Category
Height
(feet AGL) Speed
(mph)
Urbanization Terrain
Antenna
Integration
Antenna
Orientation
Min Max
GAV 0 40k 920 Urban/Suburban/Rural
Flat/Sloped/Canyon
Open/Impeded
Land/Water
Yes/No Variable
GLN 0 1,000 600 Urban/Suburban/Rural
Flat/Sloped/Canyon
Open/Impeded
Land/Water
Yes/No Variable
HPR 0 20,000 180 Urban/Suburban/Rural
Flat/Sloped/Canyon
Open/Impeded
Land/Water
Yes/No Variable
TIM 0 1000s 100 Urban/Suburban/rural
Flat
Open
Land
No Fixed
CEL 0 100s 100s Urban/Suburban/rural
Flat/Sloped/Canyon
Open/Impeded
Land/Water
Yes Variable
SPB 1,700k 4,300k 16k n/a n/a No Variable
1
In an effort to further down-select representative use cases, priorities identified by the space-
Based PNT EXCOM and PNT Advisory Board were compiled as a method to prioritize the use
case development. These priorities include:
• Existing use cases
• Vital to economic, public safety, scientific, and national security
• Focus on HPR and TIM
• Focus analysis on most sensitive cases
• Apply the 1 dB degradation criteria
• Include Multi-GNSS

80
When factoring in these priorities, three use case scenarios were identified for further in-depth
investigation:
1) Agriculture/Farming
2) Construction/infrastructure
3) Emergency response
A graphic of the emergency response uses case is shown in Figure 4-4. First responders are
increasingly using GPS to locate patients both during emergencies and as a normal course of
duty. As shown in the figure, there are multiple use of GPS in this scenario applying GLN
receivers for navigation of the emergency service response vehicles, as well as asset tracking,
including increased awareness of where response personnel and vehicles are located. A GAV
receiver on a drone also plays a role in this scenario, supporting the response effort. Drones are
becoming of increasing importance in collecting imagery and sensor data in response to natural
disasters and other incidents.
This scenario illustrates that use of a GPS/GNSS receiver can be quite close in distance (10’s of
meters) to a base station transmitter and potentially very close to a handset as well transmitting in
the adjacent band. The GPS/GNSS receiver also potentially could be vertically above the base
station height.
Figure 4-4: Emergency Response Use Case

81
A presentation given at workshop VI can be found in Appendix H, which provides a breakdown
of the three use case scenarios. Table 4-3 summarizes the geometric parameters of the three
priority scenarios. These applications/use cases happen routinely and bound the impact of base
station transmitters.
Table 4-3: Summary of Geometric Parameters
Use Case
Scenarios
Use case Category Vital Needs Lateral
Distance(s)
Vertical
height(s)
Agriculture/
Farming
Precision
Farming
HPR Economic
10 ft and
greater from
base station
0-20 ft above
ground
Crop Health
Monitoring
GLN/GAV Economic
10 ft and
greater from
base station
Up to and
above base
station
Construction/
Infrastructure
Surveying HPR Economic
1 city block
and greater
from the base
station
Up to and
above base
station
Emergency
Response
UAS/UAV GLN/GAV
Public
Safety/
National
Security
10 ft and
greater from
base station
Up to and
above base
station
Emergency
Services
GLN/CEL
Emergency
Response
Receiver Antenna Patterns
Models for GLN, GAV, TIM, HPR, and CEL receivers relative antenna gain patterns as a
function of frequency were developed based upon the antenna characterization activities
described in Section 3.3.
The following simple model was found to be representative for relative VPOL and HPOL
antenna gain patterns for GLN, GAV, TIM, and HPR receivers:
2
( )
G θ αθ
= −
where G is the relative antenna gain (dBi), ɵ is the off-boresight angle (deg), and α is a unitless
coefficient.
Based upon curve fits using the 14 external, active antennas that were measured in an anechoic
chamber as discussed in Section 3.3, the coefficients in Table 4-4 were determined. The curve
fitting approach used was a standard unweighted linear least squares fit of the single parameter
equation above.

82
Table 4-4: Coefficients for GLN, GAV, TIM, and HPR Receivers for Modeling Relative
VPOL Antenna Gain at 22 Frequencies*
Frequency (MHz) α, GLN&GAV α, TIM α, HPR
1475 3.6511e-04 6.6446e-04 8.2449e-04
1490 4.0306e-04 7.4609e-04 8.4546e-04
1495 4.5153e-04 7.4928e-04 8.4870e-04
1505 4.6656e-04 7.4815e-04 7.6944e-04
1520 4.9953e-04 7.6698e-04 7.6808e-04
1530 4.9687e-04 7.4564e-04 7.7055e-04
1535 4.4305e-04 7.4764e-04 7.5991e-04
1540 7.0113e-04 7.7206e-04 7.5869e-04
1545 6.5594e-04 7.5573e-04 7.7657e-04
1550 5.0195e-04 6.8500e-04 8.1978e-04
1555 5.4545e-04 6.3767e-04 8.5491e-04
1575 5.7732e-04 5.5176e-04 8.5922e-04
1595 5.3406e-04 6.0901e-04 8.6792e-04
1615 3.9454e-04 5.0824e-04 8.2166e-04
1620 4.2042e-04 5.4509e-04 8.2117e-04
1625 4.5397e-04 5.4762e-04 8.1460e-04
1630 4.7544e-04 6.6388e-04 8.2114e-04
1635 4.2583e-04 6.3971e-04 8.3291e-04
1640 3.5254e-04 5.5736e-04 8.3908e-04
1645 3.4695e-04 5.4974e-04 8.4719e-04
1660 4.4364e-04 5.8069e-04 7.8310e-04
1675 4.7622e-04 5.9775e-04 8.4784e-04
*In the table entries, “e-04” denotes an exponent to the minus 4 power, i.e., “×10-4
”.
As an example of the relative gain patterns, Figure 4-5 shows the modeled relative VPOL
antenna gain patterns at 1530 MHz.

83
Figure 4-5: Relative VPOL Antenna Gain Patterns for 1530 MHz
The coefficients were generated using the following procedure:
• The 14 measured antennas were grouped by category (GLN/GAV, TIM, or HPR).
• Within each category, and for each frequency, the VPOL antenna patterns were:
o Adjusted by estimated active subassembly gain at L1 so that they nominally
included only passive element gain and filtering.
o Converted from dBi to linear units, averaged, and then converted back to dBi.
• The mean VPOL antenna pattern for each category and each frequency was then:
o Forced to be symmetric with off-boresight angle from -180 to 180 deg
o Fitted with a quadratic polynomial. Since only the relative pattern is of interest,
the bias term is not important. The forced symmetry results in the linear term
being equal to zero. The quadratic term became the α value within Table 4-4.
This procedure addresses the following considerations:
• Given that only a small set of measured patterns were available for each category, the raw
data averaged across units within a category includes variations that would not be
expected from a larger sample size.
• When averaging patterns, the pattern for the antenna with the least amount of attenuation
at each frequency was deemed to be most important (because an antenna with a
tremendous amount of filtering would be associated with a receiver with a high ITM that
-80 -60 -40 -20 0 20 40 60 80
-7
-6
-5
-4
-3
-2
-1
0
Gain
(dBi)
GLN&GAV
TIM
HPR

84
is not greatly impacted by adjacent band interference at that frequency). This prompted
the averaging of gains in linear units (not dB).
• Asymmetries in gain patterns with positive vs negative off-boresight angle is unimportant
since the antenna could be oriented arbitrarily in azimuth relative to an adjacent-band
interference source in the real world.
Figure 4-6 provides an example of the data processing. Three GLN VPOL antenna gain patterns
at 1530 MHz are shown in the figure. These patterns already have the active subassembly gain at
L1 removed from them (15.0 – 19.9 dB for these units). The three patterns are averaged together
in linear units, converted back to dBi, and forced to be symmetric with respect to off-boresight
angle to form the “Mean” gain shown in the figure. The final curve shown in the figure is the
quadratic polynomial fit (obtained with MATLAB polyfit), with the α value shown in Table 4-4
of 4.9687e-04. Using the equation above, this model yields a relative gain value of zero at
boresight (by definition) and a relative gain value of –(4.9687×10-4
)(90)2
= -4.0 dBi at 90 deg
off-boresight angle.
Figure 4-6: Gain Patterns Illustrating Generation of GLN Coefficients for 1530 MHz
The results for HPOL are shown in Table 4-5. For base stations using +/-45 deg cross-
polarization, the VPOL and HPOL gain patterns were averaged.
-200 -150 -100 -50 0 50 100 150 200
-55
-50
-45
-40
-35
-30
-25
Gain
(dBi)
GLN Antenna 1
GLN Antenna 2
GLN Antenna 3
Mean
Poly Fit to Mean

85
Table 4-5: Coefficients for GLN, GAV, TIM, and HPR Receivers for Modeling Relative
HPOL Antenna Gain at 22 Frequencies*
Frequency (MHz) α, GLN&GAV α, TIM α, HPR
1475 4.8398e-04 5.5084e-04 8.4574e-04
1490 4.7233e-04 6.2100e-04 8.3577e-04
1495 4.8102e-04 6.2399e-04 8.3705e-04
1505 5.1078e-04 6.4098e-04 7.6026e-04
1520 5.8403e-04 6.7548e-04 7.4805e-04
1530 6.5353e-04 6.6919e-04 7.4907e-04
1535 7.1505e-04 6.7349e-04 7.3921e-04
1540 6.4548e-04 6.8364e-04 7.4000e-04
1545 5.4709e-04 6.7359e-04 7.6239e-04
1550 5.6432e-04 5.0510e-04 8.1090e-04
1555 5.5046e-04 4.4691e-04 8.4732e-04
1575 4.5639e-04 4.6423e-04 8.5749e-04
1595 5.0855e-04 5.5868e-04 8.5147e-04
1615 6.5552e-04 5.4371e-04 7.9655e-04
1620 6.4930e-04 5.4973e-04 7.9416e-04
1625 6.6186e-04 5.3681e-04 7.8823e-04
1630 6.9139e-04 5.9523e-04 7.9844e-04
1635 7.6854e-04 4.6848e-04 8.0766e-04
1640 7.7504e-04 1.3496e-04 8.0420e-04
1645 7.4623e-04 3.9051e-05 8.0746e-04
1660 7.1712e-04 9.1810e-05 7.5459e-04
1675 5.9731e-04 1.0261e-04 8.5621e-04
*In the table entries, “e-04” denotes an exponent to the minus 4 power, i.e., “×10-4
”.
For CEL antennas, a relative gain value of 0 dBi is recommended for all directions for two
reasons:
• Cell-phone antennas are typically low-gain, but with erratic patterns depending on the
shielding of the cell-phone case, other components, and interaction with the human body
holding it (see, e.g., measurements in Section 3.3.3).
• The cell-phone antenna could be oriented in any direction.
Propagation Models
Three propagation models were considered within this report for all receivers except for certified
aviation and spaceborne. These models are free-space, two-ray, and the Irregular Terrain Model.
The free-space and two-ray models were introduced in Section 4.1.1. The Irregular Terrain
Model is an implementation (with improvements) of the Longley-Rice propagation model by the
NTIA1
. Propagation losses yielded by this model (on a median level) differ by less than 2 dB
from free-space propagation loss (FSPL) for the relevant distances and frequency range (i.e.
distances up to half the interspacing distance between transmitters and frequencies between 1475
and 1675 MHz). However, NTIA additionally recommends using a blended model that is FSPL
for small distances and transitions to Irregular Terrain Model starting at a 100 m distance.
1
See https://www.its.bldrdoc.gov/resources/radio-propagation-software/itm/itm.aspx.

86
Therefore when the tolerable EIRP levels are considered for distances of 100 m or less, as is the
case for civil receivers use cases (excluding use cases for space-based and certified aviation
receivers), both the blended and FSPL models yield the same results.
The equations used to perform forward and inverse modeling analysis are first developed for the
case of free space path loss propagation. The modified equations for the case of a two-ray path
loss are subsequently presented. All of the forward and inverse modeling results presented in
this report do not consider OOBE and thus the impact is dictated by the fundamental emissions
of the interference source. If an adjacent band system were deployed for which this assumption is
invalid, lower EIRP values may be necessary to protect GNSS and would need to be determined
for the applicable OOBE limits. See Section 3.2.5.2 for a discussion of OOBE levels.
For receiver power calculations, an LTE signal with dual +/-45o
polarization is equivalent to a
signal radiating with twice the power and +45o
polarization when signals in the two polarizations
are uncorrelated. This signal can then be decomposed into vertically and horizontally polarized
signals. Considering this along with propagation loss, and receiver and transmitter antenna gains,
the corresponding voltage complex amplitude received by an RHCP antenna is shown below
𝑨𝑨𝒗𝒗(𝒓𝒓
�⃗𝑻𝑻, 𝒓𝒓
�⃗, 𝒇𝒇) = �𝟐𝟐.
𝑷𝑷(𝒇𝒇)
𝟐𝟐
.
𝟏𝟏
𝑳𝑳𝑷𝑷(𝒓𝒓
�⃗)
. 𝑮𝑮𝑻𝑻𝑻𝑻(𝒓𝒓
�⃗). 𝑮𝑮𝑹𝑹𝑹𝑹𝑹𝑹(𝒓𝒓
�⃗, 𝒇𝒇). 𝒆𝒆𝒋𝒋
𝟐𝟐𝟐𝟐
𝝀𝝀
𝑹𝑹𝒅𝒅 (4-1)
𝑨𝑨𝒉𝒉(𝒓𝒓
�⃗, 𝒇𝒇) = 𝒋𝒋. �𝟐𝟐.
𝑷𝑷(𝒇𝒇)
𝟐𝟐
.
𝟏𝟏
𝑳𝑳𝑷𝑷(𝒓𝒓
�⃗𝑻𝑻,𝒓𝒓
�⃗)
�⃗). 𝑮𝑮𝑹𝑹𝑹𝑹𝑹𝑹(𝒓𝒓
�⃗, 𝒇𝒇) . 𝒆𝒆𝒋𝒋
𝟐𝟐𝝅𝝅
𝝀𝝀
.𝑹𝑹𝒅𝒅
(4-2)
𝑨𝑨(𝒓𝒓
�⃗, 𝒇𝒇) = 𝑨𝑨𝒗𝒗 + 𝑨𝑨𝒉𝒉
= �
𝑷𝑷(𝒇𝒇)
𝑳𝑳𝒑𝒑(𝒓𝒓
�⃗)
�⃗). 𝒆𝒆𝒋𝒋
𝟐𝟐𝟐𝟐
𝝀𝝀
.𝑹𝑹𝒅𝒅
. � �𝑮𝑮𝑹𝑹𝑹𝑹𝑹𝑹(𝒓𝒓
�⃗, 𝒇𝒇) + 𝒋𝒋. �𝑮𝑮𝑹𝑹𝑹𝑹𝑹𝑹(𝒓𝒓
�⃗, 𝒇𝒇) �
= �𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚�
1
2. �
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑓𝑓)
. 𝐿𝐿𝑝𝑝(𝑟𝑟
⃗𝑇𝑇, 𝑟𝑟
⃗)
. 𝑔𝑔𝑇𝑇𝑇𝑇(𝑟𝑟
⃗). 𝑒𝑒
𝑗𝑗
2𝜋𝜋
𝜆𝜆
.𝑅𝑅𝑑𝑑
. � �𝑔𝑔𝑅𝑅𝑅𝑅𝑅𝑅(𝑟𝑟
⃗) + 𝑗𝑗. �𝑔𝑔𝑅𝑅𝑅𝑅ℎ(𝑟𝑟
⃗) �
(4-3)
where:
𝑅𝑅𝑑𝑑 = |𝑟𝑟
⃗ − 𝑟𝑟
⃗𝑇𝑇| is the distance between the transmitter and receiver antennas,
𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚 is the antenna gain at boresight assumed equal for both polarizations,
𝑔𝑔𝑇𝑇𝑇𝑇(𝑟𝑟
⃗) is the normalized transmitter gain in the direction of the receiver antenna,
𝑔𝑔𝑅𝑅𝑅𝑅𝑅𝑅(𝑟𝑟
⃗) and 𝑔𝑔𝑅𝑅𝑅𝑅ℎ(𝑟𝑟
⃗) are the normalized receiver antenna gains in the direction of
the transmitter antenna for the case of horizontal and vertical polarizations
respectively.
The term 𝐿𝐿𝑝𝑝(𝑟𝑟
⃗) is the free space path loss factor defined in the following equation:

87
𝐿𝐿𝑝𝑝(𝑟𝑟
⃗) = �
4𝜋𝜋𝑅𝑅𝑑𝑑
𝜆𝜆
�
2
(4-4)
The power coupled into the receiver from its antenna output is found by taking ½ the amplitude
squared as follows
𝑃𝑃(𝑟𝑟
⃗, 𝑓𝑓) =
1
2
𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚.
⃗𝑇𝑇,𝑟𝑟
⃗)
⃗). �� 𝑔𝑔𝑅𝑅𝑅𝑅𝑅𝑅(𝑟𝑟
⃗, 𝑓𝑓) + 𝑗𝑗. �𝑔𝑔𝑅𝑅𝑅𝑅ℎ(𝑟𝑟
⃗, 𝑓𝑓) � �
2
=
1
2
𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚.
⃗)
⃗). �𝑔𝑔𝑅𝑅𝑅𝑅𝑅𝑅(𝑟𝑟
⃗, 𝑓𝑓) + 𝑔𝑔𝑅𝑅𝑅𝑅ℎ(𝑟𝑟
⃗, 𝑓𝑓)� (4-5)
This power is outside the receiver tolerance when it exceeds the tolerance level 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓). 𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚.
This inequality is shown in the equation below.
𝑃𝑃(𝑟𝑟
⃗,𝑓𝑓)
𝐺𝐺𝑅𝑅,𝑚𝑚𝑚𝑚𝑚𝑚
=
1
2
.
⃗)
⃗, 𝑓𝑓)� > 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) (4-6)
The forward modeling uses this inequality to determine the impacted region for a predetermined
EIRP(f) level.
On the other hand, the inverse modeling solves for the value EIRP (𝑟𝑟
⃗, 𝑓𝑓) that ensures
compatibility for a given transmitter and receiver location. An EIRP map for all receiver
locations in a vertical plane relative to a fixed transmitter is created using the following equation
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑟𝑟
⃗, 𝑓𝑓) =
⃗). 𝐼𝐼𝑇𝑇𝑇𝑇(𝑓𝑓)
1
2
𝑔𝑔𝑇𝑇𝑇𝑇(𝑟𝑟
⃗, 𝑓𝑓)�
(4-7)
This map is performed in the vertical y-z plane containing the phase center of the transmit
antenna and in the direction of maximum gain of a sector antenna and in any direction in the case
of an omnidirectional antenna.
For given use cases associated with one category of receivers, the range of GNSS application
operational heights and a standoff distance 𝑑𝑑𝑠𝑠 will determine the an EIRP mask 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑠𝑠, 𝑓𝑓)
according to the following equation
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑠𝑠, 𝑓𝑓) = 𝑀𝑀𝑀𝑀𝑛𝑛𝑟𝑟
⃗𝜖𝜖𝑹𝑹𝒖𝒖𝒖𝒖
{ 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑟𝑟
⃗, 𝑓𝑓)} (4-8)

88
Where 𝑅𝑅𝑢𝑢𝑢𝑢 is the use cases analysis region as shown in Figure 4-7.
Figure 4-7: Illustration of use case analysis region
In order to ensure compatibility with all receivers within a category the bounding ITMs are used.
This will result in EIRP values that protect 100% of the receivers tested. Results for the median
ITM within each category are also presented resulting in power levels that would leave 50% of
the receivers unprotected.
The inverse modeling modified equation for the case of two ray path loss is shown below.
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑟𝑟
⃗, 𝑓𝑓) =
𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓)
1
2
⃗, 𝑓𝑓) . 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,𝑣𝑣(𝑟𝑟
⃗, 𝑓𝑓). 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,ℎ(𝑟𝑟
⃗, 𝑓𝑓)�
(4-9)
Where 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,𝑣𝑣(𝑟𝑟, 𝑓𝑓) and 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,ℎ(𝑟𝑟, 𝑓𝑓) represent respectively the vertical and horizontal
polarization path factors (also sometimes referred to as path gains in the literature) derived in
Appendix F.
The inverse modeling equation can be easily rearranged to get the inequality expression for the
forward modeling
1
2
𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑓𝑓). 𝑔𝑔𝑇𝑇𝑇𝑇(𝑟𝑟
⃗, 𝑓𝑓) . 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,𝑣𝑣(𝑟𝑟, 𝑓𝑓) + 𝑔𝑔𝑅𝑅𝑅𝑅ℎ(𝑟𝑟
⃗, 𝑓𝑓). 𝑃𝑃𝐹𝐹2𝑅𝑅𝑅𝑅𝑅𝑅,ℎ(𝑟𝑟, 𝑓𝑓)� > 𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓) (4-10)
When solving for tolerable EIRP for the case of multiple transmitters, the inverse modeling
equations stay the same with a summation in the denominator over all transmitters. For example
the FSPL inverse modeling equation takes the following form when aggregate effects are
considered.
𝒅𝒅𝒔𝒔
Use case analysis region (𝑹𝑹𝒖𝒖𝒖𝒖)
TX antenna
Rx antenna
𝒓𝒓
�⃗
𝒀𝒀
𝒓𝒓
�⃗𝑻𝑻
𝒁𝒁

89
𝐸𝐸𝐸𝐸𝐸𝐸𝑃𝑃(𝑟𝑟
⃗, 𝑓𝑓) =
𝐼𝐼𝐼𝐼𝐼𝐼(𝑓𝑓)
∑ 𝛼𝛼𝑇𝑇. �
1
2
.
1
⃗)
⃗, 𝑓𝑓)��
𝑇𝑇
(4-11)
𝛼𝛼𝑇𝑇 is a power control term for the general case when EIRP is not same for all transmitters. The
analysis in this report will use 𝛼𝛼𝑇𝑇 = 1 for all transmitters whenever aggregation is considered.
Forward Modeling Results and Sensitivity
Appendix I provides a comprehensive set of forwarding modeling results for two adjacent-band
LTE base station deployment types (macro urban and small cell outdoor/micro urban; see Table
4-1) and also for LTE mobile devices. Type 2 (10 MHz LTE) signals were assumed. Base station
results were produced for each of the 11 potential adjacent-band LTE downlink frequencies
listed in Table 3-6 (1475, 1490, 1505, 1520, 1525, 1530, 1535, 1540, 1545, 1550, 1675 MHz)
and handset results for each of the 7 potential adjacent-band LTE uplink frequencies (1620,
1625, 1630, 1635, 1640, 1645, 1660 MHz).
An example of the base station forward modeling results is shown in Figure 4-8. The results
shown on the plot assumes a macro urban base station (EIRP = 59 dBm/sector, height = 25 m,
other characteristics as described in Section 3.4.2.1) operating at 1530 MHz, free space
propagation, and the most sensitive (bounding) GAV GPS/GNSS device category processing
GPS C/A-code signals. The three contours in the plot depict the two-dimensional areas where the
received interference level from a macro urban base station exceeds three thresholds: (1) the ITM
(i.e., where the bounding GAV device category experienced a 1 dB CNR degradation, (2) LOLL,
the loss-of-lock threshold for low-elevation angle satellites, and (3) LOLH, the loss-of-lock
threshold for high-elevation angle satellites.
The base station is situated near the bottom left of the plot with its antennas located at the point
(0, 25 m) in x, y where x is the lateral distance from the base station and y is the height above
ground. The GAV DUT was assumed to have an antenna with the relative gain pattern modeled
as discussed in Section 3.4.3.1.
Figure 4-8: Macro Urban Base Station (EIRP = 59 dBm), Bounding GAV, 1540 MHz

90
Appendix I also includes summary charts such as shown in Figure 4-8. Each summary chart
shows, for each applicable frequency, the maximum impacted lateral distance for each DUT type
and each LTE transmitter type (e.g. macro urban base station, small cell outdoor base station,
handset). For instance, at 1540 MHz Figure 4-9 has three data points that correspond to the
maximum horizontal extent of the impacted region contours from Figure 4-8.
Figure 4-9: Maximum Impacted Lateral Distance for Bounding GAV, Macro Urban Base
Station (EIRP = 59 dBm)
Results for the HPR receiver category at 1530 MHz are presented in Figure 4-5. Note the HPR
category experienced a 1 dB (or greater) CNR degradation beyond 14 km from the transmitter
and loss of lock occurred on low elevation satellites out to 3 km with loss of lock on all satellites
out to approximately 1 km.

91
Figure 4-60: Macro Urban Base Station (EIRP = 59 dBm), Bounding HPR, 1530 MHz
Appendix I is organized as follows. For GPS C/A-code and assuming free-space propagation,
Section I.1, I.2, and I.3 present forward modeling results for macro base stations, small cell
outdoor/micro urban base stations, and handsets, respectively. Section I.4 examines the
sensitivity of the results to:
• Less sensitive DUTs – results for the median-performing vs the most-sensitive DUTs.
• GNSS signals – results for other GNSS signal types vs. GPS C/A-code.
• Propagation models – the variability of the results with propagation model.
Inverse Transmit Power Calculation Results and Sensitivity Analysis
Inverse Transmit Power Calculation Results
Inverse modeling is used to determine EIRP tolerance masks 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑠𝑠, 𝑓𝑓) for a category of
GNSS receivers and for a given standoff distance ds. The details of this analysis including the
relevant equations and parameters are described in section 4.1.4 . This inverse modeling is only
applied to the 10 MHz LTE downlink frequencies. The treatment of uplink frequencies is only
considered in the forward analysis since at the time of the writing of this report the authors were
not aware of any proposals to limit the maximum EIRP for handsets to a value below what is
specified in the M.2292 document.
For the HPR category, the EIRP map for the bounding L1 C/A ITM is shown in Figure 4-11 (a)
for the single micro urban base station at a center frequency of 1530 Hz. The use case analysis
has shown that receiver heights extends to at and above the height of a base station in all
categories and therefore the tolerable EIRP as a function of standoff distance can be found by
taking the minimum along heights up to and above base station heights as shown in Figure 4-11
(b). The extent of the impact region is >10 km from the transmitter for an EIRP of 29 dBW and
1.8 km for EIRP of 10 dBW.
0 2000 4000 6000 8000 10000 12000 14000
Distance from Tower (m)
0
1000
2000
Height
(m)
≥ 1 dB C/N
0
degradation
Loss of Lock on Low Elevation Satellites
Loss of Lock on All Satellites

92
Figure 4-71: Tolerable EIRP results. (top) Tolerable 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝒓𝒓
�⃗, 𝒇𝒇) map in the vertical
computation domain, (bottom) Tolerable 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) as a function of standoff distance X.
However, the computation domain only needs to extend up to half the distance between the
nearest two base stations in a uniform network deployment. For the case of micro urban
deployment a computation domain up to 500 meters is sufficient. Below is the zoomed in version
of the Figure 4-12 to illustrate the EIRP levels that protect HPR receivers processing L1 C/A
signals at short distances from the transmitter.

93
Figure 4-82: Tolerable 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) as a function of standoff distance X up to X=500 m
The marker in this figure indicates the maximum tolerable EIRP, 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑑𝑑𝑠𝑠, 𝑓𝑓) , with ds=10 m
and f=1530 MHz. If this is repeated across all base station frequencies, a maximum tolerable
EIRP mask can be generated as shown in Figure 4-13 below.
0 50 100 150 200 250 300 350 400 450 500
X(m)
-80
-75
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
Tolerable
EIRP
(dBW)
Tolerable EIRP vs. Distance from Transmitter inside Z=[0,30]
X: 10
Y: -41.08

94
Figure 4-9: EIRP(f,ds=10m) for the HPR category: L1 C/A, micro urban deployment,
bounding EIRP Mask, and FSPL propagation
It is worth noting that use cases indicated that receivers can be as close as 10 ft (3.0 m) to the
base station. It was not clear that they can approach that distances at heights comparable to that
of the base station. If receivers approach the base station height at the 10 ft standoff distance the
tolerable EIRP levels will be lower by approximately 5 dB. This analysis can be repeated to
generate EIRP masks for all categories of receivers at different standoff distances. Figure 4-14
depicts the L1 C/A EIRP masks for all five categories for the 10 m standoff distance. A more
comprehensive set of results for 10, 100 and 500 m standoff distances that protects L1 C/A along
with all other emulated GNSS signals are shown in Appendix J.
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
15
20
EIRP
(dBW)
L1CA, Micro Urban, Bounding EIRP Mask: FSPL, d
Standoff
=10m
HPR

95
Figure 4-10: EIRP(f,ds=10m) for five receiver categories of receivers:
L1 C/A, micro urban deployment, bounding EIRP Mask, and FSPL propagation
The HPR tolerable EIRP levels that protects L1 C/A and all GNSS signals for the same 10 m
standoff distance are compared in Figure 4-15.
Figure 4-11: Comparison of EIRP(f,ds=10m) L1 C/A and All GNSS masks for the HPR
category of receivers: Micro urban deployment, bounding EIRP Mask, and FSPL
propagation
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
15
20
EIRP
(dBW)
Standoff
=10m
GAV
GLN
HPR
TIM
CEL
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-70
-60
-50
-40
-30
-20
-10
0
10
20
EIRP
(dBW)
AllGnss, Micro Urban, Bounding EIRP Mask: FSPL, d
Standoff
=10m
L1CA
AllGnss

96
As expected, the levels that protect all GNSS signals are lower than the ones that protect L1 C/A
signals since they are calculated based on the minimum of all bounds across emulated services.
Figure 4-16 presents the resulting all GNSS EIRP masks for five categories of receivers at a
standoff distance of 10 m.
Figure 4-12: EIRP(f,ds=10m) for five categories of receivers: All GNSS, micro urban
deployment, bounding EIRP Mask, and FSPL propagation
An exhaustive list of plots that include results for the macro deployments are shown in Appendix
J.
Sensitivity Analysis
The transmit power level results presented in Section 4.1.6.1 considers only a single transmitter
and FSPL propagation model. This subsection examines how these results vary when
aggregation effects of multiple transmits are considered. The sensitivity of these results to the
propagation model used is also considered. Finally, the transmit power levels corresponding to
the median ITMs are also discussed are part of this sensitivity analysis.
Aggregation Effects
When multiple transmitters are radiating at equal EIRP, the single base station is expected to
dominate for small standoff distances. As this distance increases, the aggregate effects become
significant and limit the tolerable EIRP levels below that of a single transmitter. The aggregation
analysis is here performed using a micro urban deployment of two full rings of adjacent cells
around the center cell. In Figure 4-17, the center cell is in white, the inner ring of adjacent cells is
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
15
20
EIRP
(dBW)
AllGnss, Micro Urban, Bounding EIRP Mask: FSPL, d
Standoff
=10m
GAV
GLN
HPR
TIM
CEL

97
in yellow, and the outer ring of adjacent cells is in green. For urban and suburban regions, signals
emitted from additional transmitters outside what is simulated here will have diminished effect
on aggregate results. Additionally, they are expected to encounter blockage from buildings and
terrain that will further diminish their contribution to the final results.
Figure 4-13: micro deployment used for the aggregation sensitivity analysis.
A small cell of radius 𝒓𝒓𝒄𝒄𝒄𝒄𝒄𝒄𝒄𝒄, and transmitters’ interspacing distance ISD.
This analysis was performed in the vertical plane for the center transmitter (i.e. y=0 plane) for
the HPR L1 C/A Bounding ITM. These results are overlaid with that of a single base station in
Figure 4-18.
𝑿𝑿

98
Figure 4-18: (a) Overlay of 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) as a function of standoff distance X for the case of
single and multiple base stations, (b) 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) ratio in dB for the two cases
As previously discussed, the computation domain was limited to half the distance between
transmitters beyond which the tolerable EIRP will start dropping again due to the proximity to
the next transmitter. Figure 4-18 (a) shows the aggregation effects to be noticeable for standoff
distances greater than 20 m. For example, a standoff distance 100 m the aggregation effect
reduces the tolerable EIRP by approximately 1.8 dB relative to the case of a single transmitter.
This reduction grows to about 5.5 dB at 200 m standoff distance.
In Figure 4-19 the tolerable EIRP(f,ds) masks for the single and multiple transmitters cases are
compared at a standoff distance of ds=100 m. It shows similar reduction in EIRP on the order of
2 dB for all frequencies.
0 20 40 60 80 100 120 140 160 180 200
X(m)
-100
-50
0
EIRP(X,f)
(dBW)
1525 MHz, L1CA, Bounding EIRP Mask, FSPL
MicroUrban 1-BS
MicroUrban Multi-BS
0 20 40 60 80 100 120 140 160 180 200
X(m)
-10
-5
0
EIRP(X,f)
Ratio
(dB)
EIRP
dBW,MicroUrban Multi-BS
(X,f) - EIRP
dBW, MicroUrban 1-BS
(X,f)
(a)
(b)

99
Figure 4-19: Overlay of 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) tolerance masks for the case of a single base station
and that of multiple transmitter case
Effects of Propagation Models
The results shown so far are based on FSPL propagation. The sensitivity of HPR results to the
use of two ray path loss as opposed to FSPL model is considered in this section for the case of
micro urban single transmitter. The Irregular Terrain Model is the same as the FSPL model for
standoff distances up to 100 m and is therefore indirectly accounted for in this analysis. The two
ray path loss tolerable EIRP map and the EIRP function of standoff distance are Figure 4-20.
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-50
-40
-30
-20
-10
0
10
EIRP
(dBW)
Standoff
=100m
MicroUrban 1-BS
MicroUrban Multi-BS

100
Figure 4-14: Tolerable EIRP levels for the case of two ray path loss propagation model.
(a) Tolerable 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝒓𝒓
�⃗, 𝒇𝒇) map in the vertical plane,
(b) tolerable 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) as a function of standoff distance X.
The two ray 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸(𝑋𝑋, 𝑓𝑓) is overlaid with the FSPL in Figure 4-21 (a), and their difference is
shown in Figure 4-21 (b). These figures show that tolerable EIRP levels are similar for both
models up to a distance of about 20 m after which the two ray path loss results in lower tolerable
levels. For a standoff distance of 100 m, the two ray path loss results in 4.8dB lower tolerable
level than that of FSPL.
HPR EIRP
Tol
Map at f o
= 1525MHz with an ITM(f
o ) = -70.4213dBm
0 100 200 300 400 500
X(m)
0
50
Z(m)
-50
0
50
0 100 200 300 400 500
X(m)
-75
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
Tolerable
EIRP
(dBW)
Tolerable EIRP vs. Distance from Transmitter inside Z=[0,30]
(a)
(b)

101
Figure 4-15: (a) Overlay of 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) as a function of standoff distance X for the case of
FSPL and two ray path loss propagation, (b) ratio of EIRP(X,f) in the above plot in dB
This analysis is applied to the remaining downlink frequencies and an EIRP tolerance mask for
the two ray path loss is produced. This mask is overlaid with that of FSPL in Figure 4-22. A
more comprehensive set of results is presented in Appendix J.
0 50 100 150 200 250 300 350 400 450 500
X(m)
-75
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
EIRP(X,f)
(dBW)
1525 MHz, L1CA, Micro Urban, Bounding EIRP Mask
FSPL
2Ray
0 50 100 150 200 250 300 350 400 450 500
X(m)
-10
-8
-6
-4
-2
0
EIRP(X,f)
Ratio
(dB)
EIRP
dB, 2Ray
(f,X) - EIRP
dB, FSPL
(f,X)
(a)
(b)

102
Figure 4-16: Comparison between two ray and FSPL EIRP tolerance masks 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝑿𝑿, 𝒇𝒇) for
X=100m standoff distance
EIRP masks for Median ITMs
EIRP levels based on median ITMs protect 50% of the tested receivers and leave the rest
unprotected. The resulting EIRP levels corresponding to the median masks are shown in Figure
4-23 for a micro urban cell transmitter and L1 C/A signals.
Figure 4-17: EIRP levels corresponding to L1 C/A median ITMs
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
EIRP
(dBW)
L1CA, Micro Urban, Bounding EIRP Mask: 2Ray, d
Standoff
=100m
FSPL
2Ray
1
5
0
5
1
5
2
0
1
5
2
5
1
5
3
0
1
5
3
5
1
5
4
0
1
5
4
5
1
5
5
0
1
5
7
5
1
6
2
0
1
6
2
5
1
6
3
0
1
6
3
5
1
6
4
0
1
6
4
5
1
6
6
0
1
6
7
5
Frq (MHz)
-70
-65
-60
-55
-50
-45
-40
-35
-30
-25
-20
-15
-10
-5
0
5
10
15
20
EIRP
(dBW)
L1CA, Micro Urban, Median EIRP Mask: FSPL, d
Standoff
=10m
Bounding
Median

103
Because of the linearity of the inverse modeling equation, at a particular frequency, the
difference in the tolerable EIRP levels equals the difference between the bounding and the
median ITMs for that same frequency in dB.
Summary of Transmit Power Level Calculation
The approach to determine tolerable EIRP levels for a given standoff distance (inverse
modeling), as well as the one to determine minimum standoff distance for a given EIRP value
(forward modeling) were described in section 4.1.4. Interference source (transmitter)
characteristics were primarily obtained from M.2292 (Characteristics of Terrestrial IMT-
Advanced Systems for Frequency Sharing/Interference Analyses) and proposals to FCC for
adjacent band network applications. Base station characteristics are summarized in Table 4-1 and
handset characteristics are summarized in Section 4.1.2.2. Base station antenna patterns are
shown in Figure 4-2 and Figure 4-3.
GNSS receiver antenna measurements for each one of the 22 frequencies used in the WSMR
tests were done to determine the appropriate antenna pattern to use for each category of
receivers. Parabolic fits to these measurements were ultimately used as inputs to the forward and
inverse modeling calculations. The results of these fits are shown in Table 3-16 and Table 4-5
for vertical and horizontal polarization respectively. The propagation loss was estimated through
the FSPL model and the Two-ray model. Since the Irregular Terrain Model is expected to have
the same properties as FSPL for distances up to 100 meters it is indirectly considered as part of
the FSPL analysis.
Tolerable EIRP levels for base stations that protect all tested receivers processing the L1 C/A
signal are shown in Table 4-6 at standoff distances of 10 and 100 meters for two different
deployments. The base station results for receivers that process the other tested GNSS signals are
shown in Table 4-7.

104
Table 4-6: Tolerable Base Station 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝒅𝒅𝒔𝒔, 𝒇𝒇) for L1 C/A bounding masks for Type-2
Interference signal using FSPL propagation model
Tolerable EIRP (dBW) by Interference Frequency (MHz)
Deployment Type 𝒅𝒅𝒔𝒔 (m) Cat
1475
1490
1505
1520
1525
1530
1535
1540
1545
1550
1675
Micro Urban
10
GAV -14.25 -10.21 -16.92 -23.37 -25.15 -29.99 -31.93 -32.06 -41.96 -51.03 -13.38
GLN -13.94 -16.9 -19.58 -23.37 -25.15 -29.99 -31.93 -32.06 -40.02 -49.38 -7.41
HPR -23.11 -28.65 -33.55 -34.55 -38.55 -41.08 -43.01 -49.75 -57.86 -61.12 -16.1
TIM 15.22 14.71 6.65 -5.44 -10.9 -19.85 -26.67 -31.24 -41.14 -50.61 12.73
CEL n/a*
n/a*
n/a*
n/a*
13.15 10.77 8.39 -2.56 -12.33 -19.85 11.26
100
GAV 5.75 9.78 3.08 -3.37 -5.15 -9.98 -11.92 -12.03 -21.95 -31.02 6.63
GLN 6.06 3.1 0.42 -3.37 -5.15 -9.98 -11.92 -12.03 -20.01 -29.38 12.59
HPR -3.03 -8.56 -13.49 -14.5 -18.49 -21.02 -22.96 -29.7 -37.8 -41.04 3.99
TIM 35.23 34.74 26.69 14.6 9.14 0.19 -6.63 -11.2 -21.1 -30.6 32.73
CEL n/a n/a n/a n/a 33.15 30.77 28.39 17.44 7.67 0.15 31.26
Macro Urban
10
GAV -14.77 -10.75 -17.52 -24.02 -25.8 -30.66 -32.59 -32.82 -42.67 -51.67 -14.02
GLN -14.46 -17.44 -20.17 -24.02 -25.8 -30.66 -32.59 -32.82 -40.72 -50.02 -8.05
HPR -24 -29.54 -34.39 -35.39 -39.38 -41.92 -43.84 -50.58 -58.7 -61.99 -17
TIM 14.51 13.93 5.87 -6.25 -11.71 -20.65 -27.47 -32.05 -41.94 -51.3 12.42
CEL n/a n/a n/a n/a 13.22 10.84 8.46 -2.49 -12.26 -19.78 11.33
100
GAV 5.22 9.24 2.47 -4.03 -5.81 -10.68 -12.6 -12.85 -22.68 -31.68 5.97
GLN 5.53 2.56 -0.18 -4.03 -5.81 -10.68 -12.6 -12.85 -20.74 -30.03 11.93
HPR -4.04 -9.58 -14.42 -15.42 -19.41 -21.95 -23.87 -30.61 -38.73 -42.03 2.96
TIM 34.5 33.91 25.85 13.73 8.27 -0.67 -7.49 -12.08 -21.96 -31.32 32.41
CEL n/a*
n/a*
n/a*
n/a*
33.22 30.84 28.46 17.51 7.74 0.22 31.33
*n/a signifies no CNR degradation of 1-dB was detected within the tested range of interference power

105
Table 4-7: Tolerable Base Station 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝒅𝒅𝒔𝒔, 𝒇𝒇) for All GNSS bounding masks for Type-2
Interference signal using FSPL propagation model
Deployment Type 𝒅𝒅𝒔𝒔 (m) Cat
1475
1490
1505
1520
1525
1530
1535
1540
1545
1550
1675
Micro
Urban
10
GAV -14.25 -27.21 -25.92 -28.37 -28.88 -32.97 -33.7 -35.01 -41.96 -51.03 -13.38
GLN -19.94 -27.21 -25.92 -28.37 -28.88 -32.97 -33.7 -35.01 -41.52 -56.83 -8.05
HPR -26.11 -33.65 -33.71 -45.08 -45 -44.82 -44.8 -51.79 -59.85 -62.2 -19.1
TIM 15.22 4.5 4.8 -5.44 -10.9 -19.85 -26.67 -31.24 -41.14 -50.61 8.23
CEL 10.68 13.7 14.52 15.35 13.15 10.68 5.25 -2.56 -21.78 -37.68 11.26
100
GAV 5.75 -7.22 -5.92 -8.37 -8.87 -12.96 -13.7 -14.98 -21.95 -31.02 6.63
GLN 0.06 -7.22 -5.92 -8.37 -8.87 -12.96 -13.7 -14.98 -21.51 -36.83 11.95
HPR -6.03 -13.56 -13.65 -25.02 -24.94 -24.76 -24.75 -31.74 -39.79 -42.13 0.99
TIM 35.23 24.53 24.84 14.6 9.14 0.19 -6.63 -11.2 -21.1 -30.6 28.23
CEL 30.68 33.7 34.52 35.35 33.15 30.68 25.25 17.44 -1.78 -17.68 31.26
Macro
Urban
10
GAV -14.77 -27.75 -26.51 -29.02 -29.53 -33.64 -34.36 -35.77 -42.67 -51.67 -14.02
GLN -20.46 -27.75 -26.51 -29.02 -29.53 -33.64 -34.36 -35.77 -42.22 -57.47 -8.69
HPR -27 -34.54 -34.55 -45.91 -45.83 -45.65 -45.63 -52.62 -60.69 -63.08 -20
TIM 14.51 3.73 4.02 -6.25 -11.71 -20.65 -27.47 -32.05 -41.94 -51.3 7.92
CEL 10.75 13.77 14.59 15.41 13.22 10.75 5.32 -2.49 -21.71 -37.61 11.33
100
GAV 5.22 -7.76 -6.53 -9.03 -9.54 -13.65 -14.37 -15.8 -22.68 -31.68 5.97
GLN -0.47 -7.76 -6.53 -9.03 -9.54 -13.65 -14.37 -15.8 -22.24 -37.49 11.3
HPR -7.04 -14.58 -14.58 -25.94 -25.86 -25.68 -25.66 -32.64 -40.72 -43.11 -0.04
TIM 34.5 23.71 24 13.73 8.27 -0.67 -7.49 -12.08 -21.96 -31.32 27.91
CEL 30.75 33.77 34.59 35.41 33.22 30.75 25.32 17.51 -1.71 -17.61 31.33
As expected from the WSMR tests on receiver susceptibility, the smallest base station EIRP is
imposed by the HPR receivers. For L1 C/A signals and macro-urban networks, the tolerable
EIRP decreases monotonically from about -24 dBW (4 mW) at 1475 MHz, to -42 dBW (< 0.1
mW) at 1530 MHz, to -62 dBW (< 1 𝜇𝜇𝜇𝜇) at 1550 MHz; for micro-urban networks the results
increase by a fraction of a dB. For all GNSS signals, the above values decrease by a few dB.
The tabulated results also show that the results are not sensitive to the deployment type when a
single base station is considered. The differences between the two deployments are ≤ 1 dB for
any frequency, category, and standoff distance combination. The average difference is 0.6 dB.
However, the levels that protect all GNSS signals can be as much as 15 dB lower than those
needed to protect L1 C/A signals from base station emissions with an average difference of 3.5
dB across all frequencies and five categories considered in Table 4-6 and Table 4-7. It is worthy
to note that the difference in results between 10 and 100m standoff distances is a constant of 20
dB with a tolerance of less than 0.1 dB despite accounting for the antenna pattern. This is
because for FSPL propagation the tolerable EIRP at a particular standoff distance is found when
the phase center of the receiver antenna is approximately aligned with the centerline direction of
the transmit antenna’s main beam. This will result in a very small difference in angles of
incidence and therefore similar receiver gain value at the 10 and 100 meters standoff distances.
Therefore, the difference in results between these two standoff distances is primarily controlled
by the difference in FSPL which is the ratio of the distances squared in dB.
These values become even smaller if two-ray path loss and aggregation effects are considered.
Also, these results did not show significant sensitivity to the transmitter antenna types (omni or
sectoral antennas associated with the deployment type).

106
Tolerable EIRP levels for handsets that protect all tested receivers processing the L1 C/A signal
are shown in Table 4-8 at a standoff distance of 10 m. The results in Table 4-8 assume free space
propagation and only a single handset. As for the base station results, the EIRP values would
become even smaller if two-ray path loss and aggregation effects are considered.
Table 4-8. Tolerable Handset 𝑬𝑬𝑬𝑬𝑬𝑬𝑬𝑬(𝒅𝒅𝒔𝒔, 𝒇𝒇) for GPS L1 C/A-code bounding masks for Type-
2 Interference signal using FSPL propagation model at a standoff distance of 10 m
Cat
1620
1625
1630
1635
1640
1645
1660
GAV -19.2 -17.1 -7.1 -3.7 -5.2 -5.2 -6.6
GLN -41.3 -38.1 -31.0 -18.1 -13.7 -14.7 -11.9
HPR -57.0 -47.1 -31.3 -28.3 -28.2 -29.8 -22.1
TIM -26.3 -19.0 -10.3 -5.9 -1.8 2.7 11.8
CEL -26.3 -18.1 0.2 9.4 10.9 12.8 13.1
4.2 Spaceborne and Science Applications
This section of the report describes the analysis and evaluation of a proposed LTE base station
network operating on adjacent radio frequency bands to space-based receivers. The emphasis of
this section is on the assessment to GNSS receivers used as a science application. Additional
information can be found in Appendix K.
The following evaluation assesses the impact to one of these GNSS-based science applications,
radio occultations (RO), where space-based GNSS receivers are used to perform measurements
of the troposphere, stratosphere, and up through the layers of the atmosphere until reaching the
ionosphere. This is not to say that the other GNSS-based science applications are not affected by
a proposed LTE base station network, but RO science is an application that is particularly
susceptible and, thus, the focus of this assessment. RO measurements of the atmosphere,
coupled with traditional methodologies for Earth observation, have significantly improved
accuracy and predictability of weather forecasts. RO measurements of the ionosphere have also
improved our ability to monitor ‘space weather’ (the distribution of charged particles in the
uppermost part of the atmosphere), which is essential to ensure the successful operation of
satellites.
Specifically, NASA’s assessment focuses on the RO receiver, called the TriG (formerly also
known as TriGNSS), which was developed by the NASA/Jet Propulsion Laboratory (JPL). The
TriG is the newest RO receiver of the BlackJack class of GNSS receivers and can perform
substantially more (up to three (3) times more) measurements than previous versions. The
increase in performance is partially due to the TriG’s ability to receive signals from all GNSS
constellations including the GPS, GLONASS, Galileo, BeiDou, regional space-based navigation
constellations such as QZSS and NavIC, and SBAS, such as Wide Area Augmentation System
(WAAS) and European Geostationary Navigation Overlay Service (EGNOS).

107
Radio Frequency Interference (RFI) is a particular problem when GNSS signals are being used
for science applications. During RO measurements, the GNSS signal is defocused by tens of dB
at low ray heights, where the signal-to-noise ratio (SNR) is already in a marginal zone. In fact,
in this already marginal zone, tracking loops cannot be closed and the captured data is running
open loop. Additional noise from RFI contaminates these marginal-SNR data over specific
areas. The spatially correlated noise can bias the captured data and greatly affect the recent
climate record, while providing incorrect weather predictions over the affected areas.
Radio Occultation (GNSS-RO)
RO/GNSS-RO is the disruption/interruption of GNSS signals from a spacecraft by the
intervention of a celestial body. RO is a relatively new method for the indirect measurement of
temperature, pressure and water vapor in the stratosphere and the troposphere. These
measurements are made from specifically designed GNSS receivers on-board a Low-Earth-Orbit
(LEO) satellite. The techniques utilize the unique radio signals continuously transmitted by the
GNSS satellites (GPS, GLONASS, Galileo, etc.) orbiting the Earth at an approximate altitude of
20,000 km above the surface. The GNSS radio signals are influenced both by the electron
density in the ionosphere and by the variations of temperature, pressure and water vapor in the
atmosphere which are used in meteorology. RO measurements are also used to derive various
ionospheric parameters (Total Electron Content (TEC), Electron Density Profiles (EDP), L-band
scintillation, etc.) for understanding earth and space weather dynamics.
From the point of view of a LEO satellite (at an altitude of 700-800 km), the GNSS satellites
continually rise above, or set behind, the horizon of the Earth. During these so-called "radio
occultation", where the GNSS and the LEO satellite are just able to "see" each other through the
atmosphere, the GNSS signals will be slightly delayed and their ray path slightly bent (refracted)
on the way through the layers of the atmosphere (see Figure 4-24). The excess range increases
as the ray propagates through denser media at lower altitudes (and highly-refractive water vapor
in the atmosphere). This delay is a function of density (n/V), which is related to temperature by
the ideal gas law: P*V = n*R*T.
A typical occultation sounding will last one (1) to two (2) minutes, and during this time the LEO
satellite will receive signals where the ray paths have different minimum distances to the surface
of the Earth, from zero up to approximately 100 km. The GNSS satellites transmit on multiple
frequencies, and with a receiver rate of 50 Hz this will yield around 6000 rays, making up two
profiles of phase residuals up/down through the lowest 100 km of the atmosphere and the
ionosphere up to, or down from, the ~700 km height of the LEO satellite.

108
Figure 4-24: Straight Line versus Actual Path of GNSS Signal
The residual positioning error and determination of time delays, derived from the measurements
taken during a RO event, are key parameters in the obtaining the temperature, pressure, and
water vapor characteristics of the atmosphere at different heights. Given sub-mm measurement
precision, RO can determine atmospheric temperature profiles to 0.1 – 0.5 Kelvin (K) accuracy
from 8 - 25 km height levels.
NASA has several radio occultation receivers in its portfolio, including the Integrated GPS
Occultation Receiver (IGOR), the IGOR+, and the more recently developed receiver called the
TriG receiver.
NASA/JPL TriG Receiver Overview
The NASA/JPL developed TriG receiver functions as a multi-function GNSS receiver. This
single receiver has multiple antenna inputs and can be configured to operate in a navigation
capacity, as well as, simultaneously, in a scientific measurement role. In its traditional function,
coupled with a choke ring antenna, the TriG serves as a device for space vehicle navigation and
precise orbit determination (POD). The receiver provides accurate information to space vehicle
operators on position, velocity, and time.
Configured in a scientific measurement mode, the TriG, coupled with a series of specially
designed antenna arrays, performs RO measurements of GNSS signals. TriG receivers are able
to receive all GNSS signals: GPS, Galileo, GLONASS, Compass, as well as other navigation
signals (QZSS, DORIS, etc.). This capability increases the number of RO measurements that
can be made during any given orbit.
Spaceborne Receiver Assessment for Science-Based Applications
NASA has performed an assessment of the potential impacts caused by a proposed terrestrial
LTE network operating in the adjacent band to GPS L1. Two (2) future science missions,
COSMIC-2 and Sentinel-6 (formerly, Jason Continuity of Service (Jason-CS)), were used as the

109
basis for these assessments. NASA’s assessment is to the TriG receiver performing a science
application using the RO technique.
To determine the impact to the TriG receiver, the aggregate interference power at the output of
the TriG receiver antenna was calculated using MATLAB to model the interference scenario, as
well as the TriG receiver system, and simulate the interference effects to the satellites in orbit.
Satellites operating in LEO gain a much broader view of the earth (dependent upon antenna
characterizations and operating parameters), which must be accounted for in performing the
analysis.
Figure 4-18: Example Satellite View of the U.S. Cities
Unlike the assessments performed in Section 3, in-orbit satellites will see a greater number of
potential interference sources (e.g. – increased number of terrestrial Base stations (ES)) and the
aggregate of those interference sources will be the major contributing factor in the assessment,
see Figure 4-25.
Summary of TriG Receiver System Characteristics Used for Analysis
Table 4-9 summarizes the satellite TriG receiver system characteristics for the analyses
performed on COSMIC-2 and Sentinel-6. The interference threshold in this table is the RFI
power at the output of the flight RO antenna which causes a -1 dB C/No degradation in the TriG
receiver as used in the COSMIC2-A mission. It was derived from the power density observed by
the 0 dBiL standard gain horn used in during the DOT ABC test at a RFI power level causing a 1
dB C/No degradation. Since the TriG choke ring antenna was located at a different spot, it
actually received about 3.2 dB more RFI power per meter squared (m2
). In addition, the choke
ring antenna had about +3.7 dBi linear gain toward the RFI source, adding 3.7 dB to the
threshold power. After these corrections, the LTE power at 1530 MHz that causes a 1 dB C/No
degradation is -78.2 dBm + 3.2 dB + 3.7 dB = -71.3 dBm, defined at the output of the receive
antenna.
Another adjustment that was made to estimate the effect on the flight receiver is the difference in
noise floors due to the extra antenna temperature from black body radiation coming from the
ceiling and walls of the WSMR anechoic chamber. During the test, the noise floor is estimated

110
to be 349 Kelvin (K). This is based on preamplifier (Preamp) noise of 51 K, antenna temp of 300
K, and filter loss of 0.8 dB. The noise floor in flight is estimated to be 224 K based on Preamp
noise of 51 K, antenna temp of 150 K, and filter loss of 0.8 dB. This difference shows an
adjustment to lower the 1 dB threshold by 1.9 dB. Therefore, the normalized in-flight RFI power
of is calculated to be approximately -73 dBm (-71.3 dBm – 1.9 dB = -73.2 dBm) from the
antenna corresponding to a -1 dB degradation of C/No.
Table 4-9: Summary Table of Satellite TriG Receiver Characteristics Used for Modeling
and Simulation
Receiver Characteristic COSMIC-2 Sentinel-6
Satellite Orbit Altitude 800 km 1330 km
Satellite Orbit Inclination Angle 72° 66°
TriG Forward Receive Antenna Type 12-Element Array 6-Element Array
TriG Forward Receive Antenna
Downtilt (relative to satellite velocity
vector)
26.2° 34.2°
TriG Forward Receiver Antenna Main-
Beam Gain @ 1530 MHz (single
subarray)
+ 13.4 dBic + 10.5 dBic
TriG Aft Receive Antenna Type Not modeled 12-Element Array
TriG Aft Receive Antenna Downtilt
(relative to satellite velocity vector)
Not modeled 34.0°
TriG Aft Receiver Antenna Main-Beam
Gain @ 1530 MHz (single array)
Not modeled + 12.5 dBic
Interference Threshold (-1 dB C/No) - 73 dBm - 73 dBm
Terrestrial LTE Deployment Scenarios
The aggregate interference is dependent upon several factors. A few factors are satellite related,
including orbital parameters and receiver system characteristics. The other determining factor
comes from the interference sources. The most important factor is the transmitter characteristics
and the total number of sources (e.g., LTE base stations (BS)). Since TriG receiver systems
(performing the RO technique) operate in LEO, they have a direct line-of-sight to a broad area of
the U.S., and the aggregate interference is dependent upon the long-term deployment scenario of
the LTE operator.

111
NASA used three parameters, the City Zone model, the City Population and the BS Cell radius to
determine the total number of BS that could be deployed in the LTE network. The assumptions
used for each of the parameters are described below.
The City Zone model was used to determine the physical area around a city center location that
the simulated LTE network would be deployed over. The baseline City Zone model was chosen
to conform to the only available accepted model given in ITU Report ITU-R SA.2325-0 [4]
(International Mobile Telecommunication (IMT) sharing at 2GHz) for an BS deployment based
on three (3) zones (e.g. – urban, suburban, and rural) with given radial distances from a city
center latitude/longitude location. An example City Zone model with the typical macro cellular
will have a hexagonal grid layout deployed about a city center. Because the LTE services to be
provided by the proposed and analyzed network may not be as widespread in terms of city area
as the conventional LTE deployment described in SA.2325-0 a second City Zone model with a
smaller Suburban and Rural zone size was analyzed. Parameters for both the City Zone models
are listed in Table 4-10.
Table 4-10: Zone Model - ES Zone-specific Radial Distance from City Center
Zone Model
Urban Zone
(km)
Suburban Zone
(km)
Rural Zone
(km)
1 0 – 3 3 – 20 20 – 50
2 0 – 3 3 – 10 10 – 30
In addition to a City Zone model it was necessary to define the BS cell radius (CR) parameter in
order to determine the BS grid layout within each City Zone. The typical M.2292 zone values
listed in Table 4-11 were used as the baseline cell radius (CR) in the simulation.
In consideration to the where the proposed LTE network is to be deployed, the size of the city
population was an additional parameter that was included in the simulations. If a U.S. city had a
population of greater than 125,000, but less than 250,000, it was included in the analyses for half
of the simulations. Cities with populations of over 250,000 were included in all simulations.
Accordingly, the number of assumed cities included in each simulation was chosen from:
• City Population > 125K: 225 cities or
• City Population > 250K: 82 cities

112
Figure 4-196: Base station Deployment Zone Model (Report ITU-R SA. 2325-0)
Table 4-11: Typical Cell Radius (CR) - M.2292
Zone Type City Population CR (km)
Urban All 0.5
Suburban All 1.0
Rural All 5.0
In addition to the above ‘typical’ model for the cell radius, half of the 16 simulation groups use a
variation of the cell model, referred to as ‘scaled’ model. In the ‘scaled’ model, the cell radius
increases up to double its typical value, as the city population decreases. This decreases the
effective number of cell stations, as well as the resulting interference.
Using the set of Zone Model, City Population and Cell Radius parameters, NASA calculated the
total number of BS required for deployment for each simulation run. Table 4-12 depicts the
number of Base stations for the set of three parameters for a LTE network deployment consisting
of only macrocells.
Table 4-12: Total # of BS (Macrocell Deployment Only)
Zone Model
City
Population
(in 1000s)
Cell Radius
Number of BS
Urban Suburban Rural Total
1 > 125 Table 4-10 11,700 143,100 29,700 184,500
1 > 250 Table 4-10 4,264 52,152 10,824 67,240
2 > 125 Table 4-10 11,700 33,750 12,150 57,600
2 > 250 Table 4-10 4,264 12,300 4,428 20,992

113
Summary of BS Transmitter System Characteristics Used for Analysis
In addition to the parameters described above, the following simulation parameters were
considered and chosen by NASA for the analysis performed.
• BS antenna side-lobe pattern:
o F.1336-4 Recommends 3.1. (Macro)
o F.1336-4 Recommends 3.2. (Micro)
Elevation Mask: Two (2) BS mask angles are utilized for the analysis:
o A 0° elevation mask on the BS so that all BS that see the satellite above 0° elevation
angle are included in the aggregate interference calculation, and
o A 5° mask angle so that only BS that see the satellite above 5° elevation angle
contribute to the aggregate interference.
BS Activity Factor (AF):
An AF of 3 dB, corresponding to 50% of the Base stations transmitting simultaneously, is
used throughout the analysis.
Note: If 100% of the Base stations are transmitting simultaneously, the peak interference
levels in the results will be 3 dB higher. In this case the other resulting interference statistics
would be increased in time duration or frequency of occurrence as well.
BS Transmitter Power (EIRP):
Table 4-13 depicts the nominal transmit power used for some of the simulations (as per
M.2292). Considerations were also given to the maximum transmit powers of 10 dBW and
32 dBW EIRP per channel per sector.
Table 4-13: Assumed Transmitter Levels per Sector (Typical per M.2292)
BS Type
Typical Max. Transmit
Power/Channel/Sector (EIRP)
Macrocell - Urban 26 dBW
Macrocell - Suburban 26 dBW
Macrocell - Rural 28 dBW
Microcell (any zone) 7 dBW
TriG Receiver Analysis
Two (2) NASA missions (COSMIC-2 and Sentinel-6) that include the TriG receiver, as a
science-based function (e.g. - RO technique) were utilized for analysis. A MATLAB simulation
program was developed to model the receiver on-board a satellite, using mission-specific
parameters, and interference statistics were calculated for an LTE network deployment of BS
distributed in U.S. cities.

114
For the spaceborne receiver analysis the aggregate interference power at the output of the GPS
receiver antenna is calculated at ten (10) second time steps in the satellite orbit from BS
distributed among U.S. cities. The MATLAB program was set up to model a 10-day orbit of the
satellite.
The analysis calculates the interference value and is not dependent upon the carrier signal. The
aggregate interference to the receiver antenna output is calculated using a summation of the
interference from each source. A simple link budget formula is used to calculate the interference
received from a single source, LTE BS. The total aggregate interference is determined through
the summation of interference from the individual sources:
Rx Int Pwragg = ∑(Int sources) Tx Pwr (EIRP) off-boresight – FSPL – Pol Loss + Rx Ant
Gain off-boresight
where,
Rx Int Pwragg = Aggregate interference power level (dBm)
Tx Pwr (EIRP) off-boresight = Tx power output including antenna off-boresight
calculations (dBm) (See below)
FSPL = Free Space Loss (dB)
Pol Loss = Loss of dissimilar polarizations (Linear to RCHP Polarization = - 3
dB)
Rx Ant Gain off-boresight = Rx antenna gain including off-boresight calculations
(dBic)
The macro and micro cell sector antenna gain value towards the satellite for each time step was
calculated by determining the off-boresight azimuth (AZ) and elevation (EL) look angle gain
value from the appropriate F.1336 model gain pattern equations2
.
A total of 96 simulation runs were performed for COSMIC-2, while a lesser, but still
representative, number of runs (16 runs) were performed for Sentinel-6. Each of the simulation
runs varied one or more LTE BS deployment parameters.
While it is unknown how the LTE operator will be performing their network deployment, the
variations in simulation runs should be demonstrative. Further, the variations in runs may be
representative of an LTE network through its various phases of deployment (initial deployment
through full deployment).
Table 4-14 shows how the various parameters of the terrestrial network and the space receiver
are modeled in the different runs of simulation-1 group.
2
As defined in M.2292 and F.1336-4.

115
Table 4-14: Summary of Simulation Runs
Run
Sim
No.
Run
Designator
COSMIC-
2
Sentinel-
6
BS Tx EIRP
Zone
Model
City
Population
Cell
Radius
Elevation
Mask
Macrocell
Only
Total # of Earth
Stations
1 1 a X
M.2292
levels
1 > 125K Typical 0° X 184,500
2 1 b X X
M.2292
levels
1 > 125K Typical 5° X 184,500
3 1 c X 32 dBW 1 > 125K Typical 0° X 184,500
4 1 d X 32 dBW 1 > 125K Typical 5° X 184,500
5 1 e X 10 dBW 1 > 125K Typical 0° X 184,500
6 1 f X X 10 dBW 1 > 125K Typical 5° X 184,500
Results
The aggregate interference results for the TriG receiver, functioning as a science measurement
instrument, are presented in the following.
The received aggregate interference levels calculated during the simulations range from -90 dBm
to -40 dBm.
The following tables use an aggregate interference threshold of -73 dBm (1526 – 1536 MHz)
which corresponds to a -1 dB degradation of receiver C/No.
It should be noted that the loss-of-lock threshold for the TriG receiver occurs between -59 to -35
dBm aggregate interference power in the 1526-1536 MHz band. Loss-of-Lock at -59 dBm was
seen in Test 04 with RFI at 1525 MHz and LOL at -35 dBm was seen in Test 04 at 1530 MHz.
The entries in the results tables are interpreted as follows:
• Column 3: Max Int. Level (dBm)
Indicates the maximum aggregate interference level calculated at the receiver antenna
output.
• Column 4: % Time > Threshold
Indicates the percent time, over the 10-day simulation period, where the aggregate
interference at the TriG receiver antenna output exceeds the threshold level (-73 dBm).
As an example, if the value is about 10% of the time, the TriG receiver will have C/No
degraded by at least 1 dB for a cumulative of 24 hours.
• Column 5: # of Int Events
Indicates that over the 10-day period, the total number of interference events which
exceed the -73 dBm threshold.
• Column 6: Avg Dur Int Event (min)
Indicates the mean average duration (in minutes) of an interference event for the entire
10-day period. As discussed before, the duration of an atmospheric occultation (as the
signal path moves from skimming the Earth‘s surface to an altitude of about 100 km) is
only one to two minutes.

116
• Column 7: Max Int Event (min)
Indicates the maximum duration (in minutes) that was recorded for a single interference
event over the 10-day period.
• Column 8: Max Allow EIRP Level (dBW/10 MHz)
Indicates a reverse-engineered maximum BS transmitter power level (in dBW)
distributed across a 10 MHz bandwidth per channel per sector. The reverse-engineered
value calculated in this column would bring the interference level to the -73 dBm
threshold value for 1 dB C/No degradation. The calculated level is based on the
maximum interference level received during the 10-day period.
Table 4-15 shows the COSMIC-2 results for the simple scenario of macro cell BS at 32 dBW
EIRP.As the number of stations decreases from simulation 1 to 2 for the zone-1 model, and from
simulation 5 to 6 for the zone-2 model, there is about 5 dB less interference in zone-2 compared
to zone-1, which is expected because the zone-2 model uses about 3 times less stations. There is
about 4 dB less interference in models using transmitter elevation mask of 5o
(run d) compared to
the 0o
mask (run c), indicating that less than half of the available stations affect the satellite in
the 5o
mask case. For the most challenging model (1c), using 184,500 macro cell stations, the
tolerable EIRP is 11 dBW.

117
Table 4-15: COSMIC-2 Interference Results (Macro BS Only, All BS Tx Power 32 dBW)
Sim
No.
Run
Designator
Max Int.
Level
(dBm)
%
Time >
Thresh
# of
Int
Events
Avg Dur
Int.
Event
(min)
Max Int
Event
(min)
Max Allow EIRP
Level (dBW/10
MHz)
1 c -52 6.9 141 6.9 14.5 11
1 d -56 4.7 101 6.5 11.8 15
2 c -57 5.0 132 5.3 12.0 16
2 d -61 3.4 96 4.9 10.2 20
5 c -57 4.5 109 5.7 12.8 16
5 d -61 3.0 74 5.8 10.2 20
6 c -62 2.9 99 4.1 10.3 21
6 d -66 1.9 55 4.8 8.2 25
Table 4-16 shows the Sentinel-6 results for the simple scenario of macro cell ES at 32 dBW
EIRP, and as the number of stations decreases from simulation 3 to 4 for the zone-1 model.
There is about 2 dB less interference in models using transmitter elevation mask of 5o
(run d)
compared to the 0o
mask (run c), indicating that more than half of the available stations affect the
satellite in the 5o
mask case. For the most challenging model (3c), using 74,612 macro cell
stations, the tolerable EIRP is 23 dBW.
Please note that simulations 3 and 4 use the above mentioned variation of the cell model, referred
to as ‘scaled’ model, in which the cell radius increases up to double its typical value, as the city
population decreases; this results in fewer stations, and less interference, compared to the
simulations 1 and 2.
Table 4-16: Sentinel-6 Interference Results
(Macro BS Only, All BS Tx Power +32 dBW/10 MHz)
Sim No.
Run
Designator
Max Int. Level
(dBm)
%
Time >
Thresh
Max Allow
EIRP Level
(dBW/10 MHz)
3 c -64 7.3 23
3 d -66 5.2 25
4 c -68 4.9 27
4 d -70 3.4 29

118
5. CERTIFIED AVIATION RECEIVER
5.1 Determination of Tolerable Interference Levels
Certified GPS, GPS/SBAS and GPS/ground-based augmentation system (GBAS) airborne
equipment will meet their performance requirements when operating within the radio frequency
(RF) interference (RFI) environment defined in appropriate Federal Aviation Administration
(FAA) Technical Standard Orders (TSOs). These technical standard orders invoke industry
Minimum Operational Performance Standards (MOPS) developed through RTCA (RTCA/DO-
229, RTCA/DO-253 and RTCA/DO-316). Sections 3.7.2 and 3.7.3 of the International Civil
Aviation Organization (ICAO) GNSS Standards and Recommended Practices (SARPs) [5] also
contain Continuous Wave (CW) and band limited noise interference levels, respectively, for
which these receivers satisfy their performance specifications and operational objectives.
This analysis addresses all receivers compliant with the requirements3
of:
• Technical Standard Order (TSO)-C145()4
, Airborne Navigation Sensors Using The
Global Positioning System Augmented By The Satellite Based Augmentation System. This
standard invokes RTCA/DO-229, Minimum Operational Performance Standards for
GPS/Wide Area Augmentation System Airborne Equipment.
• TSO-C146(), Stand-Alone Airborne Navigation Equipment Using The Global Positioning
System Augmented By The Satellite Based Augmentation System. This standard invokes
RTCA/DO-229, Minimum Operational Performance Standards for GPS/Wide Area
Augmentation System Airborne Equipment.
• TSO-C161(), Ground Based Augmentation System Positioning and Navigation
Equipment. This standard invokes RTCA/DO-253, Minimum Operational Performance
Standards for GPS/Local Area Augmentation System Airborne Equipment.
• TSO-C196(), Airborne Supplemental Navigation Sensor for Global Positioning System
Equipment Using Aircraft-Based Augmentation. This standard invokes RTCA/DO-316,
Minimum Operational Performance Standards for GPS/Aircraft-Based Augmentation
System Airborne Equipment.
• TSO-C204(), Circuit Card Assembly Functional Sensors using Satellite-Based
Augmentation System (SBAS) for Navigation and Non-Navigation Position/Velocity/Time
Output. This standard invokes RTCA/DO-229, Minimum Operational Performance
Standards for GPS/Wide Area Augmentation System Airborne Equipment.
• TSO-C205(),Circuit Card Assembly Functional Class Delta Equipment Using The
Satellite-Based Augmentation System For Navigation Applications. This standard invokes
RTCA/DO-229, Minimum Operational Performance Standards for GPS/Wide Area
Augmentation System Airborne Equipment.
• TSO-C206(), Circuit Card Assembly Functional Sensors using Aircraft-Based
Augmentation for Navigation and Non-Navigation Position/Velocity/Time Output. This
3
Where specifications are referenced, the latest version is assumed.
4
“()” encompasses all versions.

119
standard invokes RTCA/DO-316, Minimum Operational Performance Standards for
Global Positioning System/Aircraft Based Augmentation System Airborne Equipment.
Note that many receivers were designed to comply with the RFI environments defined within
these standards even though they were certified to an earlier standard (TSO-C129a5
). This
analysis does not specifically address receivers that comply only with TSO-C129a.
However, that category of receivers6
was designed to be lower-performance and narrowband. If
the receivers assessed under this analysis are shown to be compatible with signals from a
network, the FAA then accepts any residual risk that some early-generation GPS receivers not
tested to RTCA/DO-229, RTCA/DO-253, and RTCA/DO-316 may experience harmful
interference.
Area of Aviation Operation
As the National Airspace System (NAS) continues the transition to Performance Based
Navigation (PBN), GNSS and its aircraft-, satellite-, and ground-based augmentation systems
(ABAS, SBAS and GBAS) will serve as the key enablers of satellite-based navigation and of
surveillance through Automatic Dependent Surveillance-Broadcast (ADS-B).
The Wide Area Augmentation System (WAAS), FAA's SBAS, providing service in North
America, was commissioned for initial operational capability in 2003. Users equipped with
certified WAAS equipment now have access to precision vertical approach at thousands of
airports given the development of Localizer Performance with Vertical Guidance (LPV)
procedures across the NAS. WAAS also provides these users with the ability to fly area
navigation (RNAV) procedures in the en route and terminal areas. Further, the FAA has
approved the use of WAAS for en route and terminal operation in the NAS without requiring any
other equipment onboard general aviation aircraft. WAAS is also an essential positioning source
for most ADS-B compliant aircraft.
GPS, with aircraft-based augmentations such as Receiver Autonomous Integrity Monitoring
(RAIM), serves a large number of users in the NAS. Air carriers and high end business users
integrate GPS/RAIM with their Flight Management System (FMS) to conduct RNAV procedures
within en route and terminal areas.
Currently, there are two public-use GBAS ground systems in the NAS providing Category
(CAT) I procedures serving airports at Newark and Houston. The FAA anticipates increased
adoption of GBAS in the near-future as aircraft OEMs continue to equip aircraft with GBAS and
a number of airports install GBAS following the successful implementation in Houston and
Newark. CAT II and III procedures are also anticipated with new or updated ground systems.
5
TSO-C129, Airborne Supplemental Navigation Equipment Using the Global Positioning System (GPS).
6
These receivers should not be confused with the “non-certified aviation receivers” addressed elsewhere in this
Report.

120
The analysis in this Report is based on the concept of an “assessment zone” (Figure 5-1) inside
of which GPS performance may be compromised or unavailable and GPS-based safety systems
will be impacted accordingly due to the elevated levels of RFI.
The derivation of the assessment zone concept was based on engineering and operational
assumptions where helicopter operations are the limiting factor. As expected, this concept
generated a number of comments and questions from the community. It is worth noting that one
rotorcraft operator stated that its pilots use visual reference within the assessment zone and the
assessment zone would have no negative impact on their operation. However, from [6], there
were unresolved concerns expressed by several, though not all, operators about the assessment
zone and its impacts to aviation operations and safety.
These concerns include: technical and human factors issues associated with re-initialization of
GPS after loss of the signal or when the signal reception is intermittent; workload and human
factors impacts on pilots to monitor and track assessment zone locations; the possibility that pilot
workload, confusion, or error could lead to aircraft inadvertently entering an assessment zone
and losing needed GPS functionality; and impacts to onboard and ground systems that are
dependent upon GPS, such as ADS B/C, or fixed-wing and helicopter terrain awareness warning
system (TAWS/HTAWS) including obstacle alerting [6].
The FAA has not completed an exhaustive evaluation of the operational scenarios in developing
this assessment zone. Further, the current analyses do not include an operational assessment of
the impact of the assessment zone in densely populated areas. For example, the risk posed to
people and property for operations such as unmanned aircraft systems (UAS) using certified
avionics may be significant as such aircraft may be required to operate within the assessment
zone.

121
Figure 5-1: Candidate Assessment Zone (Not to Scale)
Tracking and Acquisition Thresholds
The tracking and acquisition performance requirements for GPS airborne receivers are defined in
FAA TSO-C145, TSO-C146, TSO-C161, TSO-C196, TSO-C204, TSO-C205 and TSO-C206.
The RFI aspects of these standards are identical. The relevant characteristics were first published
in 1996 and invoked by the FAA in May of 1998. The same requirements have been harmonized
internationally [ [5], paragraph 3.7.4] since 2001. The passband for this equipment is from
1565.42 MHz to 1585.42 MHz.
Receiver Tracking Limit Criteria for Adjacent-Band RFI
MOPS adjacent- and in-band RFI rejection requirements are specified for continuous wave (CW,
narrowband) radio frequency interference for the GPS band. All TSO (and European TSO
[ETSO]) approved equipment is designed and tested to ensure that these requirements are
satisfied. For convenience, the CW susceptibility limit curve for receiver tracking mode is shown
in Figure 5-2. The adjacent-band susceptibility limits will be applied in the RFI impact analysis
of the broadband wireless handset and base station emissions. Adjacent band base station
broadband emission RFI effects are modelled as if the entire fundamental emission power is
concentrated at the emission center frequency.7
7
This assumption was validated during previous activities performed in 2011.

122
Figure 5-2: CW Interference Susceptibility vs. Frequency, Tracking Mode
To preserve the aeronautical safety margin, the maximum mean aggregate RFI power must be
kept at least 6 dB below the curves at any center frequency point8
. An additional constraint on
the aggregate RFI is that the probability the received RFI exceeds a value 2 dB below the limit
curve is less than 10-6
/hour. The 10-6
/hour probability represents a 1/10 portion of the overall
continuity requirement for aircraft operations from en route to non-precision approach9
. This 10-
6
/hour limit is understood as the probability of a single disruptive RFI event. As with previous
analyses, the frequency point for limit determination is the emission center frequency. For any
aircraft attitude under study, the aggregate mean and rare (10-6
) limits apply simultaneously. A
limit computed at one center frequency can be converted to the corresponding limit at a different
center frequency by using Figure 5-2 and the appropriate mask slope. For example, the slope
between 1525 MHz and 1565.42 MHz is -2.6843 dB/MHz.
Receiver Acquisition Limit Criteria for Adjacent-Band RFI
Another consideration is the ability for the aviation receiver to acquire GPS satellite signals.
Acquisition is normally accomplished prior to takeoff and, under ideal circumstances, GPS
acquisition is not necessary during flight. However, power interruptions on the aircraft or loss of
GPS due to aggregate RF interference require that the aircraft be capable of GPS acquisition
while airborne. Since acquisition is more demanding than tracking, the receiver standards
8
This safety margin applies for aircraft airborne and ground operations.
9
The reliability of the positioning service is specified in terms of continuity (see Section 2.3.3 of the WAAS
Performance Standard [13]) The more stringent requirement is for en route through non-precision approach where
the service is defined from the surface of the earth to 100,000 feet. The associated continuity requirement is 10-5
per hour.
-130
-120
-110
-100
-90
-80
-70
-60
-50
-40
-30
-20
-10
0
10
1500
1505
1510
1515
1520
1525
1530
1535
1540
1545
1550
1555
1560
1565
1570
1575
1580
1585
1590
1595
1600
1605
1610
1615
1620
1625
1630
1635
1640
FREQUENCY - MHz
INTERFERENCE
LEVEL
-
dBm
40 MHz
CENTERED AT 1575.42 MHz
-89.5
(1525,-12)
(1626.5,+8)
-120.5
CW INTERFERENCE
BELOW LOWER LINE
INTERFERENCE
VARIES
WITH BANDWIDTH
BELOW -89.5 dBm
1 MHz
(1618,-12)
AIRCRAFT WITH SATCOM
ALL OTHERS

123
specifications require operation with a 6 dB lower interference test condition than in the tracking
case. As a result, the acquisition test threshold is -34.1 dBm (-64.1 dBW) and applying a safety
margin would then result in an interference threshold at -70.1 dBW.
Rather than apply this limit directly, the FAA previously determined in the 2012 Interim FAA
Study Report [7] that the analysis should account for a maximum probability of 10-3
that the
interference exceeds -64.1 dBW. While this approach discounts the additional risk to acquisition
that occurs during banking or other real-world effects, it does recognize that acquisition would
likely become possible at some point as the aircraft continues to fly out of the area of peak
interference. For the assessment in this Report this particular threshold was not the limiting
condition, so for all the certified aviation use cases/interaction scenarios in this analysis only the
tracking mode was considered.
Receiver Tracking Limit Criteria for Broadband Handset RFI In-band to GPS
In this Report, all the scenarios associated with new broadband handset unwanted RFI to
certified GPS aviation receivers assume operation in the presence of a baseline non-aeronautical
noise-like RFI environment within the receiver passband (i.e., in-band RFI to the receiver). As
stated in [8], the in-band susceptibility for broad bandwidth non-aeronautical RFI is specified
(e.g., in RTCA DO-229 Appendix C, Table C-2 [9]) as -110.5 dBm/MHz (-140.5 dBW/MHz) in
a ±10 MHz band centered on 1575.42 MHz. As with the adjacent band susceptibility, this limit
represents an airborne receiver test condition limit and, for aviation safety considerations, the
mean environment aggregate RFI power spectral density (PSD) must be kept at least 6 dB below
the test limit. Recent studies (e.g. [8]) have shown that an existing baseline environment results
in an aggregate received RFI whose probability distribution tail essentially reaches the
operational probability limit for precision approach. As such, any additional aggregate impact
from new broadband wireless source unwanted emission will need to be well below that of the
baseline environment. The limit used for these analyses is that the aggregate effect from
additional in-band RFI does not increase the exceedance probability by more than 6% [10].
Transmitter and Receiver Component Assumptions
The transmitter portion is intended to be a single description for the full DOT ABC study.
Regarding the receiver, the primary assumptions are the interference threshold (above) and the
receiver antenna gain model. However, the “transmitter” material in this section describes
important assumptions used in the FAA certified aviation receiver analyses; some of which may
be different than in the other DOT analyses.
GPS Receive Antenna Gain
An FAA Federal Advisory Committee, RTCA Special Committee (SC-159), has developed a
representative lower hemisphere antenna gain pattern model for the GPS receive antenna
mounted on the top of the aircraft fuselage. The vertical and horizontal polarization pattern
models are assumed to be azimuthally symmetric and dependent solely on the elevation angle
from the aircraft horizon and represent the maximum gain for the particular RFI signal

124
polarization. The gain pattern model is dependent on the approach category for which the aircraft
is certified.
The lower hemisphere aircraft receive antenna pattern model in terms of gain versus elevation
angle (angle between the aircraft horizon and the line joining aircraft and RFI source) is
illustrated in Figure 5-3. This pattern is used for the broadband handsets and base stations
unwanted emission analyses when those source antenna heights are below the aircraft antenna
height.
Figure 5-3: Lower Hemisphere Installed V-pol and H-pol Receive Antenna Patterns Max.
Gain vs. Elevation Angle
(Cat.I GVPOL = -10 dBi for -90° ≤ el <-30°; = -10 + (5 + el/6) for -30°≤ el ≤ 0°)
(Cat. II/III GVPOL = -13 dBi for -90° ≤ el <-45°; = Cat. I GVPOL for -45°≤ el ≤ 0°)
(GHPOL = -16 dBi for -90° ≤ el <-30°; = -16 + (5 + el/6) for -30°≤ el ≤ 0°)
The upper hemisphere aircraft installed receive antenna maximum gain pattern model for linear
vertical polarization is shown in Figure 5-4. This pattern is used in cases where the source
antennas are at, or above, the height of the aircraft antenna.
-17
-15
-13
-11
-9
-7
-5
-90 -80 -70 -60 -50 -40 -30 -20 -10 0
Linear
Gain
(dBi)
Elevation Angle (Deg.)
Cat. 1 Max Gain (V-pol)
Cat. 2/3 Max Gain (V-pol)
Cat. 1 Max Gain (H-pol)

125
Figure 5-4: Upper Hemisphere Installed V-pol. And H-pol. Receive Antenna Patterns Max.
Gain vs. Elevation Angle
(GVPOL = 0 dBi, 75°≤ el; = -0.5+0. 0077(el-10), 10°≤ el ≤ 75°; = -5+0.45•el, 0 ≤ el < 10°)
(GHPOL = GVPOL, 45°≤ el ≤ 90; = GVPOL – (6*(45-el)/45), 0°≤ el ≤ 45°)
In the analyses that follow, the aircraft antenna is either assumed to be boresighted at zenith (for
an aircraft in level flight) or banked (for a banking aircraft) at a particular angle towards a
particular azimuth bearing.
Broadband Wireless Base Station and Mobile Handset Characteristics
Broadband Wireless Base Station Characteristics
The broadband wireless base stations used in this analysis are assumed to utilize a 3-lobed
transmit antenna pattern with a narrow beam elevation plane shape and a broader beam azimuth
plane shape. The three lobes are assumed to be centered nominally 120° apart and down-tilted
slightly in elevation (see Appendix G for antenna pattern details). The base station signal
radiation is assumed to be either vertically polarized or +/-45 degree cross-polarized. This cross-
polarized signal is equivalently modeled for GPS RFI analyses as a dual, equal amplitude
vertically and horizontally polarized signal. The antenna lobes are each assumed to transmit an
equal effective isotropic radiated power (EIRP) relative to beam-center. Determination of that
EIRP value for compatibility with aviation GPS operation is the goal of the study. The necessary
emission bandwidth is assumed to be 10 MHz at a 1531 MHz center frequency though other
possible center frequencies were considered.
-11
-10
-9
-8
-7
-6
-5
-4
-3
-2
-1
0
0 10 20 30 40 50 60 70 80 90
Elevation Angle (Deg)
Linear
Gain
(dBi)
V-Pol Max Gain
H-Pol Max Gain

126
The analyses used two different strategies for the key base station parameter of antenna height
above terrain: one, a fixed height for all antennas; or two, a set of heights specified by a
representative deployment over a wide area. The antenna towers were either at specific
deployment locations (e.g., a hexagonal grid of locations with fixed grid spacing) or a random
distribution of locations with a given average surface concentration.
Broadband Wireless Mobile Handset Characteristics
In order to perform the certified avionics assessment versus broadband wireless mobile handsets
operating above 1610 MHz, a worst-case approach was used. Broadband wireless handsets in
these analyses are assumed to have a similar necessary emission bandwidth as the base station
but with a center frequency at or above 1616 MHz. Maximum fundamental power was assumed
to be less than 1 Watt (0 dBW). The assumed handset antenna height above terrain is 1.8 meters
unless otherwise noted and has an omnidirectional antenna pattern. The handset is assumed to
have a specified unwanted effective isotropic radiated power spectral density limit (less than -95
dBW/MHz) within the GPS receiver band (1565.42 – 1585.42. MHz). For one scenario, ground-
based handsets were assumed to be randomly distributed with an average surface concentration
of up to 180 per square kilometer. Other scenarios utilized a different distribution.
5.2 Transmit Power Level Calculations
The following material discusses various scenarios and conditions used for the analyses in this
Report. Table 5-1 summarizes these activities.
Table 5-1: Analysis Scenarios and Conditions
Scenario Conditions
Inflight Aircraft / Ground-
based Handset
Final Approach Fix (FAF) Waypoint (WP)
Cat. I Decision Height (DH)
Cat. II DH
Inflight Aircraft / Ground
Base Station
Random and discrete tower locations,
Specified aircraft locations and altitudes,
flight attitudes: -25o
, 0o
banking
Inflight Aircraft / Onboard
Handset
10K ft altitude
Aircraft on Ground /
Onboard Handset
Aircraft antenna at 4 m
Aircraft at Gate / Single
Handset Source on or near
Boarding Stairs or Jetway
0 dBW @ 1616 MHz

127
Aircraft at Gate/Users
Inside Airport
Random distribution of thirty handsets
TAWS / HTAWS Scenarios
with Ground-based Mobile
Broadband Handsets
Three handset surface concentrations(30, 75,
180 per sq. km), with -95 dBW/MHz in the
GPS L1 receiver passband,
Two aircraft antenna heights (25.9 & 53.3 m)
TAWS and HTAWS
Scenarios with Broadband
Base Station
Base stations located on a grid with 433m or
693m inter-station distance. Base station
heights of 6, 10, 15 and 25 m were
considered, with 2, 4, 6, and 8 degree antenna
down tilt. Aircraft was assumed at the worst-
case location on the assessment zone, both
level flight and 25 degree bank toward the
base station10
. Additional parameters
including sloping ground were utilized as part
of a sensitivity analysis as described in
5.3.3.8.
Use Case/Interaction Scenario Development
The certified aviation assessment considered five use cases or interaction scenarios. In all
scenarios, the key parameters of interest were aircraft and source antenna heights and orientation,
number and relative location of the sources with respect to the aircraft, and the aircraft GPS
receiver operation under assessment. The results for these scenarios are summarized below and
additional information can be found in the FAA GPS Adjacent-Band Compatibility Study
Methodology and Assumptions with RTCA SC-159 [10].
Inflight Aircraft/Ground-Based Source Scenario Set
As noted above, for all the interaction scenarios the GPS receivers are assumed to operate in the
signal tracking mode. Four sub-cases were considered within this set: Handsets, discretely-
located base stations, randomly-located base stations and TAWS/HTAWS/low-altitude
scenarios.
Inflight Aircraft/Ground-Based Handset Cases
The geometric parameters for this group of cases (Final Approach Fix waypoint (FAF WP),
Category I decision height and Category II decision height) were developed from previous
studies. The FAF WP case was also used to represent airborne terminal area operations, while
the other 2 cases represent limiting cases on aircraft precision approaches. The mobile broadband
10
These parameters focus on a “small cell” topology for the broadband wireless base stations.

128
ground-based handsets in these cases were assumed to have a 1.8 meter antenna height and
randomly located in a uniform distribution at one of three different surface concentrations (30,
75, 180 per sq. km) extending to the radio horizon (except where excluded from annular sector
zones). The assumed unwanted EIRP level for these handsets was -95 dBW/MHz in the GPS L1
receiver passband.
Inflight Aircraft/Ground-Based Base Station Cases
The interference analysis methodology for the ground base station cases used a representative
scenario encompassing three different aircraft waypoint locations (JTSON, WIRSO, FIROP) on
the RNAV (RNP 0.11) approach to DCA Runway 19. Corresponding antenna heights (548.6 m,
125.64 m, 67.52 m) were used to represent points on a typical aircraft approach to a landing. The
aircraft was either in level flight or in a 25° bank toward the worst-case direction. These cases
were assessed under discrete and random base station location scenarios.
Inflight Aircraft / Discretely-located Ground Base Station Cases
For the discretely located case, base stations are at different radii and typically have a varying
height distribution. In the discrete propagation model the effective antenna height of a given
tower was generally the tower height above the ground at its base (taken from representative
deployment data) added to a correction term that accounts for the average height of the tower
base above mean sea level (MSL). The local ground height at the tower base was determined
from The National Map of the United States Geological Survey (USGS). This data is available
at “https://nationalmap.gov/elevation.html”. The aircraft antenna height was adjusted for the
same average base ground height. This correction feature accounted for the first order effect the
terrain variation on the path loss and also provided accurate antenna pattern angles needed for a
“flat earth” analysis. Additional correction was used for situations where the terrain exhibits a
significant slope in the direction toward the aircraft in addition to undulation.
Inflight Aircraft / Randomly-located Ground Base Station Cases
The randomly located base station case is included in the analysis only for comparison with
results from the discrete scenario. Based on prior analysis [10], and even though this case may
under-bound the resultant power emission limit computation, these results serve as a check on
the discrete case result. Randomly located analysis was also used to address the relative impact
for higher concentrations of base stations with correspondingly smaller radius cells.
TAWS/HTAWS and Low Altitude Positioning and Navigation Scenarios
The encounter scenarios for TAWS and HTAWS are premised on aircraft operations at low
altitude relative to the terrain while using the installed GPS receiver to determine
position/velocity data for comparison with a terrain and obstacle data base. The aircraft may be
in level flight or banking up to a given angle (aircraft- and operation-dependent).
The same TAWS/HTAWS encounter geometries were also assumed to hold for low altitude
aircraft Positioning/Navigation (Pos/Nav) operations. The principal difference in Pos/Nav

129
operations is that the GPS receiver position/velocity output is used to determine aircraft flight
control signals (e.g., a helicopter on a point-in-space approach) or attitude determination (e.g.,
UAS attitude and heading reference system [AHRS] applications).
TAWS / HTAWS and Pos/Nav Scenarios with Ground-based Mobile
Broadband Handsets
In a previous analysis [7], the mobile broadband handset aggregate unwanted emissions were
determined to be most significant for the Cat II DH scenario where the aircraft antenna was 25.9
m above the ground. In that analysis, assessment zones were assumed where mobile handsets
could NOT be operated (e.g., within the airport runway object-free area, obstacle clearance zone,
etc.).
For this analysis the mobile broadband handsets were assumed to be randomly distributed at one
of 3 different surface concentrations (30, 75, 180 per sq. km). Their assumed unwanted emission
level was -95 dBW/MHz in the GPS L1 receiver passband. At these surface concentration
values, the fundamental emission effects were insignificant. The two different aircraft antenna
height cases analyzed were 25.9 m and 53.3 m.
TAWS and HTAWS Scenarios with Broadband Base Stations
The hexagonal cellular system for this scenario consists of a central tower plus 19 concentric
hexagonal rings of towers, all at a particular inter-site distance (ISD) (i.e., distance between
towers) for a total of 1,141 towers with a grid maximum radius of 8.2 km. The aircraft (in this
case a helicopter) is assumed to 250 feet (76.2 m) from the central tower at an azimuth bearing of
30 degrees. This is the same azimuth as that of the main lobe of one of the three antennas on the
central tower, the three being equally spaced 120 degrees apart. Transmissions are assumed to
be equal power vertically and horizontally polarized so both the vertical and horizontal
polarization attenuation curves of the aircraft GPS antenna were used. Both flat ground and
sloping ground scenarios were examined. The nominal emitter antenna down tilt was 6 degrees.
Handset Sources on Board Aircraft
Onboard Handset Operation for Aircraft Inflight
In this scenario, the broadband wireless handsets were assumed to be operating with an on-board
WiFi access point when the aircraft is above 10,000 feet (AGL) altitude. The handsets were
expected to exhibit similar unwanted emissions in the GPS L1 band as in their wideband
communication mode on the ground. Emissions in the WiFi transmit band (2.45 GHz) were
expected to be similar to a standard mobile WiFi transceiver. If that assertion is correct, then
these handsets would not present a special RFI compatibility issue on the aircraft where WiFi
device operation is already permitted.

130
Onboard Handset Operation for Aircraft on Ground
In contrast to the inflight scenario, when the aircraft is taxiing toward the gate the onboard
broadband handsets were assumed to communicate through a standard ground base station
outside the aircraft. Because of the partial shielding provided by the aircraft fuselage, the
handsets were assumed to operate at full transmit power for their necessary emission. The
aircraft antenna height was assumed to be 4 m above ground and at a representative location at
the start of the taxiway and the aircraft GPS receiver was assumed to be in the signal tracking
mode. Propagation of handset emissions to the aircraft GPS antenna were characterized by the
model in RTCA/DO-235 [5.3-3] Appendix E.6.2.
For the unwanted emission analyses, the GPS receiver was assumed to operate in the presence of
a baseline level of RFI emanating from other randomly-distributed sources outside the aircraft.
The analysis uses a 3x3 cabin configuration (i.e., three seats per window) meaning that there are
three seats per window location on each side of the aircraft, resulting in a total of 189 seats. The
handsets were distributed in a random assortment of discrete locations throughout the passenger
cabin for a few representative values of total handset count. Path loss values at possible locations
were taken from [5.3-3] Appendix E, Table E-10.
Aircraft at Gate Scenarios
Aircraft at Gate / Single Handset Source on or near Boarding Stairs or Jetway
This scenario used a single broadband wireless handset operating at full emission power and the
signal propagation was assumed to be far field free-space. Handset location relative to the GPS
aircraft antenna was assumed such that the receive antenna gain was -5 dBi. Given the
propagation conditions and a single source, the result is deterministic. In this case for a single
handset with 0 dBW EIRP operating at 1616 MHz, the minimum handset antenna separation
distance for compatibility is 3.5 m. This separation might be assured by aircraft fuselage size and
geometry. The unwanted handset RFI analysis was also included in the baseline RFI effect as
well as the effect of unwanted RFI from a concentration of general sources inside the airport
terminal.
Aircraft at Gate/30 Users Inside Airport
This scenario was comprised of 30 wireless broadband handsets operating in an airport terminal
gate area that generate RFI to a GPS receiver on an aircraft located outside the terminal in front of
the gate area. The scenario is well documented in [10]. The key factors for this analysis were as
follows.
1) The aircraft GPS antenna height is assumed to be 4 meters above ground and 34
meters from front edge of terminal area.
2) The handset antenna heights are all 3 m above the aircraft antenna level (2 m above
terminal floor).

131
3) The terminal area is assumed to be symmetrically spaced in front of the aircraft with a
20 meters average depth and 50 meters width.
4) 30 handsets are assumed to be uniformly distributed throughout the 1000 sq. m. area.
5) Handsets are assumed to be operating in the 1610-1656.5 MHz band with -95
dBW/MHz unwanted EIRP in the GPS L1 band.
6) The median path loss model was two-ray free-space at these distances but with
additional building loss incorporated as follows: 20% of handsets incur an additional
20 dB loss, 60% an additional 15 dB loss, and 20% an additional 10 dB loss (excess
loss assigned relative to decreasing distance from front terminal wall).
Propagation Models
The RFI propagation path loss models used for the certified aviation assessment are based on the
flat-earth approximation. In other words, the ground under the aircraft is assumed to be
essentially smooth and flat out to a radio horizon from the point on the ground directly under the
aircraft. In line-of-sight propagation conditions at radio frequencies near that of the GPS carrier,
this radio horizon value generally depends on the aircraft GNSS and RFI source antenna heights
and the amount of atmospheric refraction along the propagation path. A 4/3 Earth radius
approximation for the refractive effect on the radio horizon is used in all propagation models.
The propagation models used in this analysis can be categorized as two different types: (1) those
scenarios where diffuse scattering, diffraction, and blockage were factors analyzed using
probabilistic path loss; and (2) clear line-of-sight scenarios which were analyzed using
deterministic free space path loss. For this assessment, the point above which free space path loss
is used generally occurs at an aircraft antenna height above ground of 550 meters. Above 550
meters, various parameter limits associated with the probabilistic models are exceeded thus
making their use problematic. Also at these aircraft heights, line-of-sight conditions generally
prevail which means that use of free space path loss was most appropriate.
Single Path Propagation Model
For free space propagation, the signal power loss over a single path is given by the well-known
inverse square law propagation model. For probabilistic propagation, the models developed by
the cellular radio community are generally applicable. These models have one feature in
common; the probabilistic nature of the path loss is very well approximated by the product of a
slow fading process and a fast fading process (as a consequence, this is also true for the single
path received interference power/power density). The slow fading process is approximated by a
log-normal distribution while the fast fading process is described by a non-central chi-squared
distribution. The log-normal component is completely determined by two parameters, µ and σ
and the chi-squared process by the parameters, L, ψ0 and ρ0. The range-dependent median path
loss between the GPS antenna and the interference source determines the primary component in
the parameter µ while the remaining parameters vary with range depending on the scenario.

132
A principal component in the slow-fade parameter, µ, is the single path-median isotropic path
loss. For this analysis, the median isotropic path loss was modelled using a continuous set of
three basic deterministic range-dependent segments. For short ranges, a two-ray path loss model
was used for distances less than the first breakpoint distance “r1.” For long ranges, a Hata-
Okumura path loss model was used for distances greater than the second breakpoint distance
“r2.” At intermediate ranges, at distances greater than r1 but less than r2, the path loss model
depended on antenna heights contained in a given scenario. A modified Erceg/Greenstein model
was used as the intermediate range model for most handset scenarios (aircraft antenna height ≤
80 m, source antenna ≤ 2 m). In all other scenarios the intermediate range path loss model used
an exponential fit between the short and long range models (log-linear interpolation (on path
loss) versus range between the r1 and r2 values). In some scenarios a moderate amount of effort
was required to determine the appropriate breakpoint distances. Additional details on path loss
models and the calculation of breakpoint values and other model parameters are provided in
Appendix F.
Aggregate Effects Model
For uniformly distributed, randomly-located interference sources, once the single path
interference characteristics have been determined, it is possible to determine analytically the
mean, standard deviation, and cumulative probability distribution associated with aggregate
received interference power. In this case, the received power from a randomly located interfering
emitter was modeled as the product of a slow fading process (log-normally distributed) and a fast
fading process having a non-central chi-squared distribution with the parameters described
above.
For sources having a known discrete distribution (i.e. the location and height parameters
associated with each source are known), two possible approaches may be used to determine the
aggregate interference power, the mean value and the cumulative distribution function. The
single source received interference power in this case is also a random variable and is described
by the product of a slow fading process and a fast fading process. Thus the aggregate
interference power, its mean value and cumulative distribution can be determined using an
analytic approach. Alternatively, it is possible to use a Monte Carlo simulation to determine both
the mean aggregate interference power and cumulative probability distribution for the discrete
source distribution case. Appendix F contains details of the aggregate statistics computation.
The analysis of received aggregate interference from handset sources assumes that handsets are
uniformly distributed over some area at an unknown random distance from the aircraft GPS
receiver. Exceptions to this assumption include scenarios where the aircraft is located at the gate
with handsets located within the terminal or on a stairway about to enter the aircraft. In these
exception cases, a discrete distribution of handsets was assumed. For interference from base
station sources, both a discrete and a random distribution of base stations were assumed.

133
Tolerable Transmit Power Calculation Results and Sensitivity Analysis
The spectrum engineering assumptions and path loss models described above were used to
perform inverse transmit power calculations. Generally, this type of calculation first aggregated
at one location the RFI from all emitters contained in the given scenario then calculated the
single common EIRP transmission limit that satisfied the tolerable RFI constraint. Both mean
based and rare event based type constraints are applicable though variations on this general
method are possible and are described below.
Tolerable Transmit Power Calculation Method Overview
Two major basic types of tolerable transmit power calculations are used in this certified aviation
receiver assessment. In the broadband base station calculation method, the station fundamental
(adjacent band) EIRP is not known a priori and is the goal of the analysis. The tolerance criteria
are simple receiver-based limits (see Sec. 5.1.2.1). In contrast for the associated broadband
wireless handsets, the unwanted handset EIRP (in-band to the GPS receiver) is assumed to be at
a specified limit and baseline in-band RFI is also present. Additionally, the tolerance criterion is
different (see Sect. 5.1.2.3) in that the growth in exceedance probability of the composite RFI is
limited to a percentage above the baseline case.
Tolerable Transmit Power Calculation Method – Base Station Cases
The transmit power calculation method for the base station cases assumed that each of the three
antenna beams on a base station tower transmit with a normalized (unity) EIRP. A mean
aggregate power factor (AF) is then computed at a desired aircraft location such as a waypoint by
combining the RFI from all base station sources using the probabilistic path loss and
probabilistic models described above (Sec. 5.2.2). In linear units, the AF is the received power
divided by the EIRP. The analytic transmit power calculation method for the base station cases
has two major steps11. The mean AF is computed first and then the CDF of the AF is computed.
The random variable Z is the normalized AF, defined as AF/(mean AF). The CDF P(z) is defined
as the probability that Z is less than or equal to z.
The corresponding mean based and rare event EIRP limits for an antenna are computed using
equations (5-1) and (5-2), respectively. These equations were derived from the information
provided in paragraph 5.1.2 and the spectrum mask information of Figure 5-2. The parameter
“Zcrit” in equation (5-2) is the argument of the AF CDF that corresponds to a threshold
exceedance probability of 10-6
.
𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀_𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵_𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 = −64.11 𝑑𝑑𝑑𝑑𝑑𝑑 − 10 ∗ 𝑙𝑙𝑙𝑙𝑙𝑙10(𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝐴𝐴𝐴𝐴) (5-1)
𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅_𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸_𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿 = −60.11 𝑑𝑑𝑑𝑑𝑑𝑑 − 10 ∗ 𝑙𝑙𝑙𝑙𝑙𝑙10(𝑍𝑍𝑍𝑍𝑍𝑍𝑍𝑍𝑍𝑍) (5-2)
The more stringent of the two results from
11
In the alternative Monte Carlo method, the mean AF and the AF CDF are determined together in a single
computation.

134
equations (5-1) and (5-2) is then the applicable limit for the particular case under study.
All the RFI calculations for the base station cases assumed the emitter has a center frequency of
1531 MHz with a 10 MHz emission bandwidth. Equations (5-1) and (5-2) are specific for a
center frequency of 1531 MHz. Examples of how to convert from the 1531 MHz based EIRP
limit calculated in this analysis to the corresponding limit at a different frequency are shown in
section 5.2.3.7.
Tolerable Transmit Power Calculation Method – Handset Cases
The method for evaluating the impact of ground based broadband wireless handsets is different
and more indirect than that of evaluating the RFI impact of base station emitters. Analysis has
shown that the fundamental emission of the broadband wireless mobile handsets, at least up to
the assumed 0 dBW maximum power and operation above 1616 MHz, are not of concern for
certified avionics. As a result, rather than determining an unknown fundamental power level of
the base stations as described above, the broadband wireless handsets are assumed to operate
with a specified unwanted emission limit (-95 dBW/MHz) within the aviation GPS receiver
passband. For certain scenarios, various values for the average number of handsets per unit area
(randomly distributed) are also assumed up to a maximum. The tolerability criterion for the
handset cases (5.1.2.3) is a limit on the percentage growth in RFI impact for the addition of new
handset sources to the baseline RFI condition. For handset cases the RFI impact is quantified by
the probability of the aggregate RFI power density exceeding the certified aviation receiver
MOPS test threshold (-140.5 dBW/MHz). Appendix F has details on computing the aggregate
RFI cumulative probability distribution.
Results for Inflight Aircraft/Ground-based Handset Cases
Details of the baseline RFI impact computation are given in [8]. In summary the baseline
condition is developed by a random distribution of sources (1.8 m antenna height) out to the
radio horizon at an average concentration of 100 per square kilometer with an individual
unwanted emission of -81.1 dBW/MHz. The limiting case baseline scenario geometry in [8] is
the Category II DH waypoint (25.94 m aircraft antenna height). Table 5 of [8] shows the
aggregate received RFI power density exceeds the MOPS test threshold (-140.5 dBW/MHz) at a
probability of 3.0144x10-4
as predicted by the generalized model.
The RFI impact of the composite of broadband wireless handsets with the baseline RFI is
analyzed in two steps. First a random distribution of only the broadband wireless handsets is
analyzed with the same scenario geometric constraints as in the baseline RFI case. The handsets
(1.8 m antenna height, 180 per sq. km. average) are assumed to emit -95 dBW/MHz unwanted
power density in the GPS receiver passband (1575.42 ± 10 MHz). The desired analysis results
are the handset-only statistics (mean and CDF). Then these statistics are combined with those of
the baseline to form the composite statistics (computation details in Appendix F).

135
The handset scenario baseline and composite statistical results are shown in two CDF (1-P(z))
curves (Figure 5-5). The dashed (baseline) curve is based on an average concentration of 100
baseline emitters per square kilometer. The solid (composite) curve is based on 280 total emitters
per square kilometer. In Figure 5-5, the Z value (x-axis point)12
at which the baseline curve
exceeds the MOPS test limit is 16.4722. The associated y-axis (1-P(z)) value is 3.0144x10-4
.
Since the composite case (baseline + handsets) mean value is somewhat higher, the composite
curve Z value is 15.34512 at the MOPS threshold and the associated probability on the red solid
curve is 3.06224x10-4
. This 1.59% probability increase from the baseline probability is below
the maximum tolerable increase of 6%. Thus, this scenario is assessed as not a critical or limiting
scenario based on the assumed handset-related parameters.
Figure 5-5: Handset Scenario Probabilities
Results for Inflight Aircraft / Ground Based Base Station Cases
Results presented here were obtained for the WIRSO case discussed in Section 5.2.1.1.2. This
specific waypoint places the aircraft nadir axis at 38.8816° North latitude and 77.046° West
longitude with the GPS antenna at an altitude of 125.64 meters above Mean Sea Level (MSL).
The aircraft was located over the Potomac River near the Tidal Basin in Washington DC.
12
As defined earlier, the x-axis parameter Z is the algebraic ratio of the aggregate power density to the mean
aggregate power density. Thus the point Z=1 corresponds the mean aggregate power density.

136
The analysis was performed assuming each interfering base station operates in the adjacent band
just below the GPS L1 band with an emission bandwidth of 10 MHz. While analyses for three
different base station center frequencies was initially intended, the results herein were completed
only for the frequency 1531 MHz. Extension to other frequencies can be performed as described
in Section 5.1.2 of this Report. The base station key operational parameters are described in
Section 5.1.3.2.
The propagation model used in the analysis was that described above in Section 5.2.2.1. As
discussed, this model incorporated a median path loss component between the GPS antenna and
the interference source which, along with the normalized base station and GPS antenna gains,
determined the log normal distribution parameter, µ . The single path median isotropic path loss,
PL(r), is composed of three range dependent segments. The WIRSO median isotropic path loss
model used a two-ray model for short ranges (r< r1), an exponential fit model for intermediate
ranges (r1 ≤ r ≤ r2), and the Hata-Okumura model for longer ranges (r> r2). The detailed
definition of these models is contained in Appendix F.
As a further refinement, the WIRSO interference scenario analysis also included a terrain
dependent slope correction factor which was incorporated into the Hata-Okumura long range
median path loss model [10]. To accurately model the scenario terrain slope, the area
surrounding the aircraft location was divided into 12 azimuth sectors of nominally 30° angular
width. (See Appendix F for additional details.)
Table 5-2 lists transmission power limits computed using the WIRSO scenario. The results of
this table were obtained using both the Analytic Statistical method and the Monte Carlo method
and there is good agreement between the results of the two methods. The Mean Power Based
EIRP Limit value in the table is based on the mean limit of -64.1 dBW at 1531 MHz while the
Rare Event Based EIRP Limit value is based on the -60.1 dBW limit. These results apply for
both flight attitudes.

137
Table 5-2: WIRSO Scenario Based Limits from Two Methods
Method Flight
Attitude
Mean Agg.
Power
Factor (dB)
Mean Power
Based
EIRP Limit
(dBW)
Zcrit
Rare Event
Based
EIRP Limit
(dBW)
Analytic
Statistical
Level Flight -97.85 33.75 3.0974 32.84
Monte Carlo Level Flight -97.89 33.79 3.0205 32.99
Analytic
Statistical
Banking,
-25 deg.
-94.41 30.31 3.3547 29.06
Monte Carlo Banking,
-25 deg.
-94.47 30.37 3.5300 28.89
The results listed in Table 5-2 include values for “Zcrit”, the Z value (as defined earlier Z is the
aggregate interference factor (AF)/mean AF) for which the probability of the corresponding CDF
curve is 1x10-6
. Figure 5-6 consists of two curves for the WIRSO banking scenario which
overlap for low values of Z but diverge at Z values of about 3.35 and higher. This figure
provides an indication of the solution sensitivity to the solution method. The ordinate of each
curve is plotted as “1 - CDF” (i.e., “1 – P(z)”) instead of as a traditional CDF for the sake of
convenience. The curves shown correspond to the bottom two rows of Table 5-2. While the
precision of the Monte Carlo results for higher Z values could be improved if more time-
consuming calculations were made, that exercise is unnecessary given Z values for 1-P(z) values
below 1x10-6
are not needed.

138
Figure 5-6: WIRSO Banking Scenario 1 – P(z) values Using Two Methods
The WIRSO scenario results were also computed using a third method. The WIRSO limits
computed using the random location method are shown in Table 5-3. A random location model
scenario result was computed for each limit type because the underlying assumptions used to
model the tower locations were adjusted.
Table 5-3: WIRSO Scenario Based Limits from Random Method
Random
Model
Scenario
Flight
Attitude
Mean Agg.
Power Factor
(dB)
Mean Power Based
EIRP Limit
(dBW)
Zcrit
Rare Event Based
EIRP Limit
(dBW)
1 Level Flight -97.45 33.35 2.497 33.37
2
Banking at
-25deg
-95.48 31.38 2.497 31.41
A comparison of the EIRP limits computed using the Analytical Statistical discrete and random
location methods (random-discrete result) is shown in Table 5-4. The comparison shows
relatively good agreement for the level flight scenario mean power based limit but there are
larger differences for the banking scenario and for the rare event based limits. These differences
arise from the same fundamental issue, i.e., the highly asymmetrical distribution of the towers

139
with respect to azimuth and distance. The analytical statistical method uses actual tower
locations while the random method used random assignment based on an approximation of the
tower locational distribution. Of more significance, the analytical statistical method computes a
cumulative distribution function directly from a characteristic function which was computed
using actual tower locations. In theory the analytical statistical method will more accurately
capture the probabilistic impact of the asymmetric tower distribution. The primary purpose of
the random method in this instance was to serve as a reasonableness check for the analytical
statistical method solution. As a result, the analytical statistical method based limits of Table 5-2
are cited as the WIRSO results.
Table 5-4: Comparison of Two Methods for WIRSO Scenario Based Limits
Flight Attitude Method Delta -
Mean Power Based
EIRP Limit (dBW)
Method Delta -
Rare Event Based
EIRP Limit (dBW)
Level Flight 0.40 -0.53
Banking at -25deg -1.07 -2.35
Results for Onboard Handset Operation for Aircraft on Ground
The locations of the handsets are the random variables selected in each realization within the
Monte Carlo simulation. This effort distributed handset locations uniform randomly throughout
the cabin and computed the normalized aggregate personal electronic device (PED) power factor
(FAGG,PED/IPLMIN) which is independent of PED EIRP. Although biasing handset locations
toward the front of the aircraft (lower path losses) where potentially more first-class passengers
would operate a handset was considered, the Monte Carlo results in Figure 5-7 suggest such a
constraint is unnecessary. The difference between the maximum and mean aggregate power
factor indicates that the handset locations need not be biased toward the front. With more than
approximately 20 handsets, the difference between the maximum and mean power factor is less
than 3 dB.
Assuming 100 of the 189 possible handsets13
are operating simultaneously indicates mean
aggregate signal loss of approximately 52 dB. Using this loss with a fundamental handset power
of 0 dBW gives a power at the aircraft antenna of -22 dBm. This fundamental power is
essentially at the aviation mask with 6 dB safety margin for the 1616 MHz
frequency. Considering the emissions level of -95 dBW/MHz, this level leads to unwanted
emissions in the GPS L1 band of -147 dBW/MHz. At the limit of all 189 handsets operating, the
13
During the scenario development, the number of users were chosen to provide an overbound/stress case to support
the stated conclusion that no interference to certified avionics is expected from handsets with the postulated
technical characteristics.

140
aggregate signal loss is 48 dB which results in slight exceedance of the aviation mask with safety
margin included. The scenario of all 189 devices operating simultaneously at exactly their
maximum levels for both fundamental and unwanted emissions is considered very conservative
so this is not deemed a limiting case. These results indicate that no further assessment was
required. If a further assessment were to be performed, it should also include computation of a
baseline RFI condition without broadband wireless handsets as stated in Section 5.2.1.3.2.
Figure 5-7: Aggregate Handset Signal Loss
Results for Aircraft at Gate/30 Users Inside Airport
Computations from RTCA DO-235B Appendix E, Equations E-1 and E-2 were again applied to
compute the aggregate power factor and normalization factor. The scenario can be visualized as
shown in Figure 5-8.

141
Figure 5-8: Aircraft at Gate with Thirty Uniformly Distributed Handsets in Terminal
Figure 5-9 shows the normalized factor as function of the number of handsets. (To compute the
aggregate RFI power, add the handset EIRP to the abscissa. For example, 30 handsets with
unwanted emissions at -95 dBW/MHz and 60 handsets14
with unwanted emissions at -81
dBW/MHz result in a mean received aggregate interference power of -145.9 dBW/MHz.) These
results indicate that no further assessment was required. If a further assessment were to be
performed, it should also include computation of a baseline RFI condition (general sources inside
and outside the terminal) without broadband wireless handsets as stated in Section 5.2.1.4.1.
14
During the scenario development, the number of users were chosen to provide an overbound/stress case to support
the stated conclusion that no interference to certified avionics is expected from handsets with the postulated
technical characteristics.

142
Figure 5-9: (1-CDF) Aggregate Power Factor
Results for Inflight Aircraft TAWS/HTAWS and Low Altitude Pos/Nav
The limiting EIRP for the emitters used in this scenario was evaluated using a different
distribution of towers and a different relative aircraft location than the evaluation at the WIRSO
waypoint described above. The hexagonal cellular system of this scenario consists of a central
tower plus 19 concentric hexagonal rings of towers, all at a particular inter-site distance (ISD) for
a total of 1,141 towers with a grid maximum radius of 8.2 km. The helicopter is assumed to 250
feet (76.2 m) from the central tower at an azimuth bearing of 30 degrees. This is the same
azimuth as that of the main lobe of one of the three antennas on the central tower, the three being
equally spaced 120 degrees apart. Transmissions were assumed to be equal power vertically and
horizontally polarized so both the vertical and horizontal polarization attenuation curves of the
aircraft GPS antenna were used. Both flat ground and sloping ground scenarios were examined.
The nominal emitter antenna down tilt was 6 degrees. Sensitivity analysis for this scenario
included varying the degree of down tilt and the aircraft distance from the central tower keeping
in mind the importance of having the helicopter located in the center of a main lobe of a central
tower antenna. Other parameter sensitivity variations included using vertical polarization-only
type transmissions, varying the ISD and “rounding out” the perimeter of the hexagonal grid
system into a circle (thereby increasing the number of towers to 1,345).
In this scenario, the dual polarization nature of the RF signal makes the calculation of the r1
breakpoint more complex. This breakpoint was set to be the closest radius at which the vertical
and horizontal polarization path losses are equal just beyond the point at which the vertical

143
polarized ray is at its critical grazing angle (at approximately 112.5 m). Also different in this
study is that the Hata r2 breakpoint was set to be 1,000 m in all scenarios.
A consistent result of all the parameter variations studied is that more that 90% of all RFI comes
from the central tower. While this is not surprising due to the distances involved, a consequence
is that the two-ray model becomes the primary path loss model, whereas in the WIRSO scenario
all towers were beyond the r1 breakpoint. The scenario primacy of the two-ray model in
combination with the sensitivity of the aircraft antenna to polarization type and elevation angle
meant that an additional level of detail to the RFI calculation procedure was needed. The two-
ray path loss calculations were modified to account for the direct and reflected rays entering the
helicopter antenna at different elevations and hence attenuated differently. The two-ray path loss
model also computes different reflection coefficients (magnitude and phase) for vertically and
horizontally polarized waves.
This scenario used a different antenna pattern than the WIRSO scenario. The transmit antenna
models for both scenarios are functions of azimuth and elevation but in this scenario the central
tower antenna oriented at an azimuth of 30 degrees had a minimum gain of -15 dB imposed for
elevations lower than -22 degrees in order to account for a lack of symmetry. Details for the
base station antennas are contained in Appendix G.
The EIRP limits computed using both the Monte Carlo and Analytic Statistical methods are
shown in Table 5-5. The results show good agreement between the two methods. Some
parameter combinations were Not Computed (NC) because they obviously would not constitute a
limiting condition and are thus rendered moot. The assumptions used to calculate the results of
Table 5-5 include a helicopter located 76.20 m (250 ft.) from the central tower with all tower
heights of 25 m for flat ground scenarios. The sloping ground scenarios assume a funnel shaped
terrain with an upward slope of 10 milliradians with the central tower at the bottom of the funnel.
In the sloping ground scenario the height above local ground for all towers remains 25 m but the
effective tower height with respect to the aircraft increases with the rising ground.

144
Table 5-5: Hexagonal Grid Power Limits Computed Using Two Methods
Scenario
Number
Inter Site
Distance
(m)
Aircraft
Bank
Angle
(deg)
Terrain
Slope
(milli-
radians)
Tower
Antenna
Down
Tilt (deg)
Aircraft
Height
(m)
Mean Based Limit
(dBW)
Rare Event Based
Limit (dBW)
Monte
Carlo
Analytic
Statistical
Monte
Carlo
Analytic
Statistical
#1 693 0 0 6 16.99 13.35 13.34 13.37 NC
#2 433 0 0 6 16.99 13.31 13.3 13.36 NC
#3 693 0 10 6 16.99 13.30 13.26 13.32 NC
#4 433 0 10 6 16.99 13.21 13.11 13.29 13.06
#5 433 25 10 6 16.99 10.28 10.27 10.34 10.18
#6 433 25 10 8 14.29 10.16 10.36 10.19 NC
Results were computed for other hexagonal grid scenarios using the Monte Carlo method. These
results are shown in Table 5-6. The results were computed using the same general assumptions
listed for Table 5-5. None of the results contained in Table 5-6 indicate a more stringent limit
than that indicated by Table 5-5.

145
Table 5-6: Hexagonal Grid Power Limits Computed Using 433m ISD Flat Earth Scenario
GPS Rx
antenna
height at
Max
EIRP (m)
Max Tx EIRP (Multiple Towers) (dBW)
No Banking 25° Banking
6m Base
Station
Tower
with 2°
Down Tilt
10m Base
Station
Tower
with 4°
Down Tilt
15m Base
Station
Tower
with 6°
Down Tilt
25m Base
Station
Tower
with 8°
Down Tilt
15m Base
Station
Tower
with 6°
Down Tilt
25m Base
Station
Tower
with 8°
Down Tilt
4 24.90 20.45 16.19 13.26 14.13 11.80
6 23.84 19.83 15.60 12.96 12.92 11.72
8 22.95 19.62 15.28 12.17 11.97 10.71
10 22.51 19.61 15.37 11.79 11.44 10.40
12 22.18 19.24 15.89 11.78 11.34 10.38
14 22.41 19.59 16.75 12.36 11.60 10.37
16 * 19.91 17.81 13.06 12.45 10.45
18 * 20.31 18.46 13.90 12.63 10.68
20 * * 19.78 14.86 13.90 11.05
22 * * 21.02 15.98 15.09 11.50
24 * * 22.53 17.31 16.36 12.10
26 * * * 18.45 * 13.09
28 * * * 19.62 * 13.96
30 * * * 20.73 * 15.25
32 * * * 22.17 * 16.58
34 * * * 23.63 * 18.03
*Not assessed
EIRP limits were also computed for a helicopter on the ground. These results are shown in Table
5-7 and none of the results contained in this Table indicate a limit more stringent that that
indicated by Table 5-5.

146
Table 5-7: Power Limits for Landed Helicopter at
Various Separation Radii from Central Tower
Separation
Radius
from GPS
Rx to
Central
Tower (ft)
Base Station
Antenna height
(m)
Base Station
Antenna
down tilt
(deg)
GPS Rx
antenna
height at
Max
EIRP (m)
Max Tx EIRP
(Multiple
Towers) (dBW)
50 25 8 4 14.59
100 25 8 4 18.31
200 25 8 4 13.82
250 25 8 4 13.26
Power limits obtained from a wide range of additional scenarios were computed using the
randomly distributed base station method. The results are shown in Table 5-8. Note that some
of the limits in this table are much lower than the 10 dBW limit recommended in this section.
All limits below 10 dBW were computed using an aircraft to tower distance of only 100 ft.
instead of the cylinder radius of 250 ft. used in Table 5-5. Further, the values in this table were
read from the minimum points along a series of curves, so the aircraft height and EIRP values do
not have the same high precision as the results reported in the other tables.

147
Table 5-8: Hexagonal Grid Scenario Based Limits
from Randomly Located Base Station Method
Aircraft
Lateral Separation
Distance to
Center Tower (ft.)
Flight
Attitude
Center
Tower
Height (m)
Aircraft
Antenna
Height (m)
Mean Power
Based EIRP
Limit (dBW)
Tower
Antenna
Downtilt
(deg)
250 Level 10 7.5 17.6 2
250 Level 15 12 14.8 2
250 Level 20 10 12.6 8
250 Level 25 12.25 14 8
100 Level 10 6 5 8
100 Level 15 14 6 2
100 Level 20 16 4 8
100 Level 25 20 4 8
250 Banking 10 8 14 2
250 Banking 15 12 to 16 12 2
250 Banking 20 14 to 16 11.5 4
250 Banking 25 20 10.4 4
100 Banking 10 8 2.8 4
100 Banking 15 14 2.5 2
100 Banking 20 16 2.6 8
100 Banking 25 24 2 2
Frequency Dependencies
Using the slope of the spectrum mask of Figure 5-2 allows an EIRP transmit power limit
computed at one frequency to be converted to an equivalent limit at a different frequency. For
example, the spectrum mask shows the permissible interference level decreasing from -12.0 dBm
at 1525.0 MHz to -103.267 dBm at 1559.0 MHz, a slope of -2.68432 dB/MHz. It then follows
that a transmit limit such as 10 dBW at 1531 MHz corresponds to a limit of -16.84 dBW at 1541
(i.e., 10 dBW + (-2.68432 x (1541 – 1531) dB)) and a limit of -43.69 dBW at 1551 MHz (i.e., 10

148
dBW + (-2.68432 x (1551 – 1531) dB)). These examples apply for frequencies between 1525
MHz and 1565 MHz where the slope of the spectrum mask is the same.
Sensitivity Analysis Results
The HTAWS case presents the most restrictive limits so solution sensitivities to various
parameters are best demonstrated using examples from this case. An important interplay
between the parameters focuses on placing the aircraft in the center of the main lobe of the RFI
emitter antenna beam. The relationship of the aircraft and the emitter main lobe varies with both
antenna heights, the degree of the emitter antenna down tilt and the distance between the
antennas. Banking (vs. level flight) also has a significant impact because the aircraft antenna
gains vary with elevation angle and banking changes the effective elevation angle. Sloping (vs.
flat) ground has a noticeable impact because the Hata path loss model contains a slope dependent
parameter and the total height of all towers, except the central one, increases with radius. The
computed limit is also sensitive to the emitter polarization because, at some elevations, the
aircraft antenna gains are larger for vertically than for horizontally polarized signals. All
analyses in this Report assume either vertically polarized radio waves or an equal power
combination of vertically and horizontally polarized waves (i.e., dual polarization).
The parameters varied during the HTAWS case study are listed below and after each parameter
type the range of values explored are listed in parentheses. The computed maximum limit is
sensitive to the following parameters and the interplay between these parameters:
• The heights of the emitter (10, 15, 20, 25 meters) and aircraft antennas (4 to 35 meters)
• The down tilt angle of the emitter antenna (2, 4, 6, 8 degrees)
• The ground distance between the two antennas (100 feet vs. 250 feet and vicinity)
• Flat ground vs. Sloping ground (upward with a 10 milliradian slope)
• Level flight vs. banking (at 25 degrees)
• Vertical vs. dual polarization (equal power vertical and horizontal polarization)
Figure 5-10, Figure 5-12, and Figure 5-14 depict how a computed mean based limit is sensitive
to various parameter changes. Figure 5-11, Figure 5-13, and Figure 5-15 show the corresponding
“1-CDF” (i.e., “1-P(z)”) curve based on parameter set of the most restrictive mean based limit of
the preceding figure. Each of these three mean based limit figures contains four curves with
different amounts of antenna down tilt. Each abscissa varies the aircraft (AC) antenna height and
the ordinate displays the resultant limit values. Figure 5-10, Figure 5-12, and Figure 5-14 show
an ordered pair of numbers for the abscissa and ordinate values that correspond to the most
restrictive mean based limit. All these figures used an assessment zone radius of 250 feet.
Calculations were also performed for a standoff radius of 100 feet, which results in lower limits
(not shown). Figure 5-10 and Figure 5-12, are for dual polarization while Figure 5-14 is for
vertical polarization. Figure 5-10 uses a 20 m height emitter while Figure 5-12 uses a 10 m
height emitter and Figure 5-14 uses a 25 m height. Figure 5-10 shows a limit of 9.948 dBW, the
lowest mean based limit computed with dual polarization. The corresponding rare event based
limit is 9.869 dBW, as computed with Equation 5.3-2 using the Zcrit value of 2.558 taken from
Figure 5-11.

149
Figure 5-14 presents an even lower mean based limit of 7.945 dBW with vertical polarization
only. The corresponding rare event based limit computed with Equation 5.3-2 using the Zcrit
value of 2.530 from Figure 5-15 is slightly lower at 7.9138 dBW. Thus Figure 5-14 and Figure
5-15 demonstrate the importance of wave polarization type on the computed limit.
The rare event limit of 9.869 dBW for dual polarization is lowest limit computed at the 250 ft.
(76.2 m) assessment zone radius. The one single limit value of 9.8 dBW cited in this Report is
derived from rounding down the computed result. This rounding allows for the additional effect
of a random distribution of base station emitters, as in [10] Section.3.5.2, that extends beyond the
central hexagonal grid sources out to the radio horizon at a decreasing surface concentration. It
is very important to note that this result assumes (equal power split) dual polarization and
highlights that a requirement for cross-polarization emissions from the base stations must be
captured in any license application or issuance. A vertical polarization (only) based limit must
be significantly lower than 9.8 dBW.
Figure 5-10: HTAWS Dual Polarization 20 m Emitter Antenna-Mean Limits
0 5 10 15 20 25 30 35
AC antenna height, m
8
10
12
14
16
18
20
22
24
Max
Tx
EIRP,
dBW
250 ft Stand-off, 20 m Tx height, -25
o
AC banking, 10 mrad sloping
2
o
downtilt
4
o
downtilt
6
o
downtilt
8
o
downtilt
(8, 9.948)

150
Figure 5-11: (1-CDF) for Most Restrictive Mean Limit Condition of Figure 5-10
Figure 5-12: HTAWS Dual Polarization 10 m Emitter Antenna-Mean Limits
0.2 1 2 3 4 5
z
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
1-P(z)
250 ft Stand-off, 20 m Tx height, 8
o
downtilt, 8 m AC height, -25
o
2.558
0 5 10 15 20 25 30 35
12
14
16
18
20
22
24
26
28
Max
Tx
EIRP,
dBW
o
2
o
downtilt
4
o
downtilt
6
o
downtilt
8
o
downtilt
(14, 12.857)

151
Figure 5-13: (1-CDF) for Most Restrictive Mean Limit Condition of Figure 5-12
Figure 5-14: HTAWS Vertical Polarization 25 m Emitter Antenna-Mean Limits
0.2 1 2 3 4 5
z
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
1-P(z)
o
o
2.712
0 5 10 15 20 25 30 35
6
8
10
12
14
16
18
Max
Tx
EIRP,
dBW
o
2
o
downtilt
4
o
downtilt
6
o
downtilt
8
o
downtilt
(22, 7.945)

152
Figure 5-15: (1-CDF) for-Most Restrictive Mean Limit Condition of Figure 5-14
Certified Aviation Receiver Analysis Results Summary
RFI degradation calculations for a variety of scenarios have been performed by the FAA in order
to determine a maximum tolerable power transmission level for usage of frequencies near the
GPS L1 band. All analyses were conducted in accordance with the procedures recommended in
the applicable RTCA [10] and FAA Technical Center [8] reports with the procedures refined
when necessary. A summary of these results from the performed analyses is provided in Table
5-9.
The “Handset” cases assessed showed these do not present a limiting case or scenario for
certified aviation receivers. The “Ground Station” analyses computed aggregate RFI power
assuming an aircraft was located at the WIRSO waypoint in Washington, DC (i.e., near Reagan
National airport over the Potomac River). This analysis used a realistic set of 1,068 towers as
well as extensive modeling of the surrounding terrain so that the impact of slope on the Hata-
Okumura path loss model could be ascertained. The “HTAWS” analyses assumed towers are
deployed in a symmetrical hexagonal grid pattern with the aircraft located 76.2 meters (i.e., 250
feet) from the central tower. Terrain modelling for this analysis assumed either flat ground or an
idealized symmetrical funnel shaped terrain with a slope of 10 milliradians in all directions.
Different transmit antenna patterns were used by the Ground Station and HTAWS analyses. In
addition to tower deployment, terrain modeling and transmit antenna patterns, another major
difference between these two analyses is signal polarization. As recommended in [10], the
Ground Station analysis set assumed all radio transmissions were vertically polarized only.
0.2 1 2 3 4 5
z
10
-7
10
-6
10
-5
10
-4
10
-3
10
-2
10
-1
10
0
1-P(z)
o
o
2.530

153
However, to evaluate the HTAWS case, the FAA RFI analysis methodology evaluated vertical
polarization only, as well as dual polarization consisting of equal power vertical and horizontal
polarized transmissions. The issue of radio polarization type is significant because the aircraft
GPS antenna gain varies according to signal elevation and polarization. The Ground Stations
analysis simulated the RFI encountered by an aircraft at an altitude of 125.64 m at an actual
waypoint over the Potomac River found an EIRP limit of 28.9 dBW. The HTAWS analysis
simulated a helicopter flying within 76.2 meters of a cellular system tower of a hexagonal grid
system dictates a significantly lower limit of 9.8 dBW.
The two cases yielded such different limits due to differences in the lateral separation distance
from the aircraft to the closest tower. The distance from aircraft nadir to the closest tower base
for the WIRSO scenario is a relatively large 1,396 m while the minimum separation distance in
the hexagonal grid scenario is 76.2 m. Assuming for a moment a simple free space path loss
model, this difference in separation distance would result in a 25 dB difference in path loss to the
closest tower. The actual delta path loss to the closest tower between the two analyses is larger
because 76.2 m is within the zone of a two-ray path loss model but no tower in the WIRSO
scenario was within the two-ray zone. In all the hexagonal grid scenarios examined the central
tower provided at least 90% of the scenario total aggregate RFI power (in some scenarios much
more). Further, the hexagonal grid scenario with an Inter-Station Distance (ISD) of 433 m had
37 towers within 1,396 m, and even a grid with an ISD of 633 m has 14 towers inside the radius
of the closest WIRSO tower. Though the evaluation of the RFI at other waypoints was
suggested by the RTCA document [10], it is unlikely that these would include an aircraft flying
within 76.2 meters of a tower.
These analyses indicate that protection of certified avionics, operating under the assumption of
the described 250 foot (76.2 m) radius assessment zone, requires that the Ground Station
transmission EIRP not exceed 9.8 dBW (cross-polarized) at 1531 MHz. This limit is obtained
from the HTAWS scenario which was found to be the most restrictive of the scenarios examined.
The limit from the Discrete Tower scenario at the WIRSO waypoint was found to be 28.9 dBW
with considerably larger Ground Station ISD. Limit values at other frequencies can be computed
as described in section 5.2.3.7 using the spectral mask slope of Figure 5-2.

154
Table 5-9: Summary of Scenarios and Findings
Scenario Conditions Comments
Inflight Aircraft /
Ground-based Handset
Final Approach Fix & Waypoint, Cat. I &
Cat. II Decision Height
Cat. II determined as most stringent
case; Assessed, <6% threshold
increase, not deemed a critical or
limiting scenario (see 5.2.3.2)
Inflight Aircraft / Ground
Base Station
Random and discrete tower locations,
Aircraft level & banking
Assessed 1531 MHz at WIRSO
location 125.64 m altitude.
Differences between 0 o
, 25o
attitude
as well as rare event attributed to
tower distributions (see 5.2.3.3)
Inflight Aircraft /
Onboard Handset
Aircraft at 10K ft. altitude Assessment premised on handset
exhibiting characteristics of WiFi at
2.45 GHz, no further assessment
required (see 5.2.1.3.1)
Aircraft on Ground /
Onboard Handset
Aircraft antenna at 4 m Assessed, not deemed a critical or
Aircraft at Gate / Single
Handset Source on or
near Boarding Stairs or
Jetway
0 dBW @ 1616 MHz Assessed, 3.5 m minimum separation
distance (see 5.2.1.4.1)
Aircraft at Gate/Users
Inside Airport
Random distribution of thirty handsets Assessed, not deemed critical or
TAWS / HTAWS
Scenarios with Ground-
based Mobile Broadband
Handsets
Three handset surface concentrations with -
95 dBW/MHz in the GPS L1 receiver
passband, Two aircraft antenna heights
Assessed, found fundamental
emission effects insignificant, no
further assessment required (see
5.2.1.2.1)
TAWS and HTAWS
Scenarios with
Broadband Base Station
Base stations located on a grid with 433 m
or 693 m inter-station distance. Base station
heights of 6, 10, 15 and 25 m were
considered, with 2, 4, 6, and 8 degree
antenna down tilt. Aircraft was assumed at
the worst-case location on the assessment
zone, both level flight and 25 degree bank
toward the base station. Additional
parameters including sloping ground were
utilized as part of a sensitivity analysis as
described in 5.2.3.8.
Fixed location base stations in
hexagonal grid with 433 m and 693
m ISDs, flat earth and funnel terrain,
aircraft lateral distances of 15.2-76.2
m, 25o
and 0o
banking. Both Monte
Carlo and Analytic Statistical
methods used for assessment (see
5.2.3.6) Assessment found HTAWS
the most restrictive scenario (see
5.2.3.8)

155
6. SUMMARY
This report describes DOT’s efforts to evaluate the adjacent band radiofrequency band power
levels that can be tolerated by GPS and GNSS receivers. The assessment described in this report
addresses transmitters in bands adjacent to the 1559-1610 MHz radionavigation satellite service
(RNSS) band used for GPS L1 signals that are centered at 1575.42 MHz.
Results from GNSS receiver testing conducted in the ARL anechoic chamber facility at WSMR
to assess their sensitivity to adjacent band interference in the range 1475 to 1675 MHz are
presented in this report for the six categories of receivers tested. The radiated GNSS signals
included GPS, SBAS, GLONASS, BeiDou, and Galileo signals. The radiated interference
waveforms included 1 MHz AWGN and 10 MHz LTE signals (referred to as Type-1 and Type-
2). The GNSS and interference signals were radiated through separate and collocated antennas as
shown in the chamber layout diagram.
The collected test data capture the performance degradation of each device through the CNR
which decreases as the interference power increases and the signal power stays fixed. In this
report, the main analysis of GNSS receiver susceptibility to adjacent band interference refers to
the interference power level at which the average CNR for a device drops by 1-dB from its
baseline (interference-free) value. The resulting interference power level vs. interference
frequency is referred to as the Interference Tolerance Mask for that device. The test data were
also used for a secondary analysis of receiver susceptibility to determine the interference power
level at which a receiver assembly loses signal tracking (referred to as Loss of Lock).
These bounding ITMs per receiver category and the GPS/GNSS antenna characteristics were the
primary inputs to use case scenario assessments to determine the maximum Effective Isotropic
Radiated Power that could be tolerated in the adjacent radiofrequency bands for each GPS/GNSS
receiver category. Space-based applications are different from other GPS/GNSS applications
considered, primarily due to the need to account for aggregation effects of multiple transmitters
visible in orbit. Although OST-R derived ITMs for space-based receivers, along with other
GPS/GNSS receiver types, OST-R deferred to NASA for assessing adjacent-band transmitter
power levels that can be tolerated for this receiver category.
The L1 C/A bounding ITM is the lowest interference power at a given frequency that resulted in
a 1 dB CNR reduction for at least one receiver in the category (for each receiver category). Most
sensitive categories are the high precision and space-based receivers. The least sensitive category
is the cellular category.
The Loss of Lock power levels for high elevation angle satellites (nominal signal power -128.5
dBm) were typically 15 – 25 dB higher than the ITM levels. The loss of Lock Interference power
levels for low elevation angle satellites (signal power -138.5 dBm) were typically 5 – 15 dB
higher than ITM levels as would be expected since the low elevation were emulated by a 10 dB
reduced power levels from the nominal signal power to account for change of receiver antenna
gain at low elevations.

156
During July 2016, 14 GNSS receivers were tested for further ABC assessment in a laboratory
setting at Zeta Associates Inc. in Fairfax, VA. The test objectives were: (1) evaluation of the
impact of adjacent-band interference on signal acquisition, (2) comparison between wired and
radiated receiver susceptibility to adjacent band interference with 1 MHz bandpass noise and 10
MHz LTE, and (3) assessment of adjacent band transmitter OOBE impacts.
The ITMs from the wired test exhibited good agreement with the radiated results when the same
active antenna was used or when the bypassed active antenna components were properly
considered in the comparison. The wired test also showed that the FCC OOBE limits (base
station and handset limits) have the potential to impact ITMs as does one entrants’ proposed
OOBE limits for handsets at separation distances less than 2 meters.
The results of these tests indicate that the 1-dB CNR degradation level is a good indicator of the
region where acquisition starts to be impacted for some receivers. This is especially noticeable
for the lower power GNSS signals emulating low elevation satellites or attenuated GNSS signal
due to foliage or other environmental factor.
The approach to determine tolerable EIRP levels for a given standoff distance (inverse
modeling), as well as the one to determine minimum standoff distance for a given EIRP value
(forward modeling) were described in the approach section. Interference source (transmitter)
characteristics were primarily obtained from M.2292 and proposals to FCC for adjacent band
network applications.
Antenna Measurements for each one of the 22 frequencies used in the WSMR tests were done to
determine the appropriate antenna pattern to use for each category of receivers. Parabolic fits to
these measurements were ultimately used as inputs to the forward and inverse modeling
calculations. The propagation loss was estimated through the Free-Space Path Loss model and
the Two-ray model. Since the Irregular Terrain Model is expected to have the same properties as
FSPL for distances up to 100 meters it is indirectly considered as part of the FSPL analysis.
Understanding GPS/GNSS receiver use cases scenarios are important so that the geometric
parameters, specifically a receiver height and lateral offset from a transmitter can be determined.
Also, it is important that use cases representative each receiver category and can provide a worst-
case scenario so most, if not all, receivers in that category are protected. In addition, use cases
are needed in conjunction with ITMs, propagations models, and transmitter scenarios to
determine what power levels can be tolerated adjacent to GPS/GNSS signals.
Use cases were compiled through substantial outreach with DOT federal partners and agencies.
Members of the working group were provided a template that contained questions related to how
their organizations use GPS/GNSS receivers to support their mission. In particular, questions
included identifying height, speed, terrain, antenna orientation and integration, and urbanization
areas.

157
The use case analysis has shown that receiver heights extends to at and above the height of a
base station in all categories and therefore the tolerable EIRP as a function of standoff distance
can be found by taking the minimum along heights up to and above base station heights. The
extent of the impact region for a high precision receiver is >10 km from the transmitter for an
EIRP of 29 dBW and 1.8 km for EIRP of 10 dBW.
In the area of impact, the behavior of the GPS/GNSS receiver can become unreliable in its ability
to meet the accuracy, availability, and integrity requirements of its intended function, impacting
safety-critical applications such as transportation, the earthquake early warning system, and
space-based missions using GPS/GNSS receivers, as well as high precision users such as
precision agriculture, machine control, and surveying.
Tolerable EIRP levels to protect all tested receivers processing the L1 C/A signal are shown in at
standoff distances of 10 and 100 meters for two different deployments. For L1 C/A signals and
macro-urban networks, the tolerable EIRP decreases monotonically from about -24 dBW (4
mW) at 1475 MHz, to -42 dBW (< 0.1mW) at 1530 MHz, to -62 dBW (<1 μW) at 1550 MHz;
for micro-urban networks the results increase by a fraction of a dB. For all GNSS signals, the
above values decrease by a few dB.
For certified GPS avionics, the FAA analyzed a number of scenarios including:
1) Inflight Aircraft with a Ground-based Handset
2) Inflight Aircraft with a Ground Base Station
3) Inflight Aircraft with an Onboard Handset
4) Aircraft on the ground with an Onboard Handset
5) Aircraft at Gate / Single Handset Source on or near Boarding Stairs or Jetway
6) Aircraft at Gate/Users Inside Airport
7) Terrain Awareness Warning System (TAWS) / Helicopter TAWS (HTAWS) Scenarios
with Ground-based Mobile Broadband Handsets
8) TAWS and HTAWS Scenarios with Broadband Base Station
The analysis is based on the concept of an “assessment zone” inside of which GPS performance
may be compromised or unavailable. In this region GPS based instrument flight rules (IFR)
operations will be restricted due to the elevated levels of RFI. Different transmit antenna patterns
were used by the Ground Station and HTAWS analyses. In addition to tower deployment,
terrain modeling and transmit antenna patterns, another major difference between these two
analyses is signal polarization. The Ground Station analysis set assumed all radio transmissions
were vertically polarized only. However, to evaluate the HTAWS case, the FAA RFI analysis
methodology evaluated vertical polarization only, as well as dual polarization consisting of equal
power vertical and horizontal polarized transmissions.
The issue of radio polarization type is significant because the aircraft GPS antenna gain varies
according to signal elevation and polarization. The Ground Stations analysis simulated the RFI
encountered by an aircraft at an altitude of 125.64 m at an actual waypoint over the Potomac
River found an EIRP limit of 28.9 dBW. The Helicopter Terrain Awareness Warning System
(HTAWS) analysis simulated a helicopter flying within 76.2 meters of a cellular system tower of

158
a hexagonal grid system dictates a significantly lower limit of 9.8 dBW. A very important
difference between the two case analyses is simply the lateral separation distance from the
aircraft to the closest tower. The distance from aircraft nadir to the closest tower base for the
WIRSO scenario is a relatively large 1,396 m while the minimum separation distance in the
hexagonal grid scenario is 76.2 m.
This limit is obtained from the HTAWS scenario which was found to be the most restrictive of
the certified aviation scenarios examined. The FAA analysis of certified aviation indicate that
protection of certified avionics, operating under the assumption of the described 250 foot (76.2
m) radius assessment zone, requires that the ground station transmission not exceed 9.8 dBW
(10W) (cross-polarized) at 1531 MHz. Based on the results of the OST-R testing and analysis of
the other categories of receivers, the transmitter power level that can be tolerated by certified
aviation may cause interference with, or degradation to, most other categories of GPS/GNSS
receivers including those used for General Aviation and drones.

159
ACRONYM LIST
ABAS
ABC
AC
ACLR
ADS-B
AF
AFSS
AFTS
AFTU
AGL
AHRS
ARL
ATC
ATIS
AWGN
AZ
CAR
CAT
CDF
CEL
CNR
COSMIC
CR
CSV
CW
CYGNSS
dB
dBi
dBic
dBm
dBW
DORIS
DOT
DUT
Aircraft-Based Augmentation System
Adjacent Band Compatibility
Aircraft
Adjacent Channel Leakage Ratio
Automatic Dependent Surveillance-Broadcast
Aggregate Factor
Autonomous Flight Safety System
Automated Flight Termination System
Automated Flight Termination Unit
Above Ground Level
Attitude and Heading Reference System
Army Research Laboratory
Ancillary Terrestrial Component
Alliance for Telecommunications Industry Solutions
Additive White Gaussian Noise
Azimuth
Certified Aviation Receiver
Category
Cumulative Distribution Function
Cellular
Carrier-to-Noise density Ratio
Constellation Observing System for Meteorology, Ionosphere and Climate
Cell Radius
Comma Separated Variable
Continuous Wave
Cyclone Global Navigation Satellite System
decibel
decibel isotropic
decibel isotropic circular
decibel-milliwatt
decibel-watt
Doppler Orbitography by Radiopositioning Integrated on Satellite
U.S. Department of Transportation
Device Under Test
DSAC
EDP
EIRP
EL
BS
ETSO
EMVAF
FAA
Deep Space Atomic Clock
Electron Density Profile
Effective Isotropic Radiated Power
Elevation
Base station
European Technical Standard Order
Electromagnetic Vulnerability Assessment Facility
Federal Aviation Administration

160
FAF
FCC
FMS
FSPL
GAV
GBAS
GEO
GHz
GLN
GM
GNSS
Final Approach Fix
Federal Communications Commission
Flight Management System
Free-Space Path Loss
General Aviation
Ground-Based Augmentation System
Geostationary
gigahertz
General Location and Navigation
General Motors
Global Navigation Satellite System
GNSS-R
GPS
GNSS Reflectometry
Global Positioning System
GRACE
GUST
HITL
HPA
HPOL
HPR
HTAWS
Hz
ICAO
ICD
IFR
IGOR
IMT
IP
IPC
ISD
ISS
ITM
Gravity Recovery and Climate Experiment
Geostationary Uplink System Type-1
Human-in-the-Loop
High Power amplifier
Horizontal Polarization
High-Precision Receiver
Helicopter Terrain Awareness Warning System
hertz
International Civil Aviation Organization
Interface Control Document
Instrument Flight Rules
Integrated GPS Occultation Receiver
International Mobile Telecommunication
Interference Power
Interference Protection Criteria
Inter-Site Distance
International Space Station
Interference Tolerance Mask
ITU-R
JPL
K
kHz
km
KPI
L1 C/A
LEO
LNA
LOL
LPV
LTE
m
M&S
MATLAB
International Telecommunications Union Radiocommunication Sector
Jet Propulsion Laboratory
kelvin
kilohertz
kilometer
Key Performance Indicator
GPS L1 Course Acquisition
Low Earth Orbit
Low Noise Amplifier
Loss of Lock
Localizer Performance with Vertical Guidance
Long Term Evolution
meter
Modeling and Simulation
Matrix Laboratory

161
MHz
MOPS
MSL
MSS
mW
NaN
NAS
NASA
NC
NCO
NDA
NET
NI
NISAR
NMEA
NOAA
NTIA
OFDM
OOBE
OST-R
megahertz
Minimum Operational Performance Standard
Mean Sea Level
Mobile Satellite Service
milliwatt
Not a Number
National Airspace System
National Aeronautics and Space Administration
Not Computed
National Space-Based PNT Coordination Office
Non-Disclosure Agreement
Networks
National Instruments
NASA-ISRO Synthetic Aperture Radar
National Maritime Electronics Association
National Oceanic and Atmospheric Administration
National Telecommunications and Information Administration
Orthogonal Frequency Division Multiplexing
Out of Band Emissions
DOT Office of the Assistant Secretary for Research and Technology
PBN
PF
POD
Pos/Nav
PRN
PSD
QZSS
Performance-Based Navigation
Power Factor
Precise Orbit Determination
Positioning/Navigation
Pseudorandom Noise
Power Spectral Density
Quazi-Zenith Satellite Service System
RAID
RAIM
RAM
RF
RFI
RHCP
RINEX
RNAV
RNSS
RO
Redundant Array of Independent Disks
Receiver Autonomous Integrity Monitoring
Radiant Absorbent Material
Radiofrequency
Radiofrequency Interference
Right-Hand Circular Polarization
Receiver Independent Exchange
Area Navigation
Radionavigation Satellite Service
Radio Occultation
RTCA
Rx
Formerly Radio Technical Commission for Aeronautics (now RTCA, Inc)
Receiver
SARPS
SBAS
SC-OFDM
SNR
SPB
SPIGAT
SWO
Standards and Recommended Practices
Satellite-Based Augmentation System
Sub-Carrier Orthogonal Frequency Division Multiplexing
Signal-to-Noise Ratio
Space-Based
Software Programmable Interference Generator for ABC Testing
Space Weather Observation

162
SWOT
TAWS
TEC
TIM
TSO
Tx
µW
UAS
USCG
USG
USGS
Surface Water and Ocean Topography
Terrain Awareness Warning System
Total Electron Content
Timing
Technical Standard Order
Transmitter
microwatt
Unmanned Aircraft System
U.S. Coast Guard
U.S. Government
U.S. Geological Survey
VPOL
VSG
WAAS
WGS
WP
WSMR
Vertical Polarization
Vector Signal Generator
Wide Area Augmentation System
World Geodetic System
Waypoint
White Sands Missile Range

163
REFERENCES
[1] DOT GPS Adjacent Band Compatibility Assessment Plan December 2012.
[2] U.S. Air Force Background Paper on Use of 1-dB Decrease in C/No As GPS Interference
Protection Criterion June 2017.
[3] ITU-R Report M.2292: Characteristics of Terrestrial IMT-Advanced systems for frequency
sharing/interference Analyses.
[4] ITU Report ITU-R SA.2325-0: Sharing between space-to-space links in space research,
space orientation and Earth exploration-satellite services and IMT systems in the frequency
bands 2025-2110 MHz and 2200-2290 MHz.
[5] ICAO SARPs Annex 10 Volume I Appendix B.
[6] Report of the Tactical Operations Committee in Response to Tasking from the Federal
Aviation Administration, December 2016.
[7] Status Report: Assessment of Compatibility of Planned LightSquared Ancillary Terrestrial
Component Transmissions in the 1526-1536 MHz Band with Certified Aviation GPS
Receivers, FAA Report PR 25, January 25, 2012.
[8] Final Report: A Generalized Statistical Model for Aggregate Radio Frequency Interference
to Airborne GPS Receivers from Ground Based Emitters (DOT/FAA/TC-14/30, Rev. 1),
August 7, 2017.
[9] Minimum Operational Performance Standards for GPS/Wide Area Augmentation System
Airborne Equipment, RTCA/DO-229.
[10] FAA GPS Adjacent-Band Compatibility Study Methodology and Assumptions with RTCA
SC-159 mark-ups, RTCA Paper No. 095-15/SC159-1040.
to Airborne GPS Receivers from Ground Based Emitters (DOT/FAA/TC-14/30),
September. 30, 2014.
[12] FAA GPS Adjacent-Band Compatibility Study Methodology and Assumptions with RTCA
SC-159 mark-ups, RTCA Paper No. 095-15/SC159-1040.
[13] Global Positioning System Wide Area Augmentation System (WAAS) Performance
Standard, October 31, 2008.

164
ACKNOWLEDGEMENTS
The authors would like to thank the following organizations for their contribution to the success
of the testing and data collection efforts:
Aerospace Corporation
Army Research Laboratory (ARL), Electromagnetic Vulnerability Assessment Facility
(EMVAF)
Federal Aviation Administration (FAA)
Federal Communication Commission (FCC)
General Motors (GM)
John Deere
MIT Lincoln Laboratory (MITLL)
MITRE Corporation
National Aeronautics and Space Administration (NASA)
National Oceanic and Atmospheric Administration (NOAA)
National Telecommunications and Information Administration (NTIA)
NovAtel
Trimble
U-Blox
UNAVCO
United States Air Force (USAF)
United States Coast Guard (USCG)
United States Geological Survey (USGS)
Zeta Associates

1
APPENDIX A
WSMR TEST DETAILS

2
TABLE OF CONTENTS
TABLE OF CONTENTS................................................................................................................ 2
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 4
REFERENCES ............................................................................................................................... 5
APPENDIX..................................................................................................................................... 6
WSMR Test Details................................................................................................ 6
A.1 Interference System Development Considerations........................................................ 6
A.2 Chamber Mapping ......................................................................................................... 8
A.3 Chamber Mapping Comparison and Stability.............................................................. 10
A.4 System Calibration....................................................................................................... 18
A.5 Interference System Linearity...................................................................................... 19
A.6 Comparison of Measured vs. Predicted Power............................................................ 21
A.7 SPIGAT Antenna Characterizations............................................................................ 23
A.7.1 AST-1507AA ........................................................................................................ 23
A.7.2 LB-510-10-C-NF................................................................................................... 27
A.8 WSMR Activity Summary........................................................................................... 29
A.9 Chamber OOBE........................................................................................................... 30
A.10 GNSS Signal Generation ........................................................................................... 38
A.10.1.1 Transmit antenna characterization ............................................................... 38
A.10.2 Almanacs............................................................................................................. 39
A.10.2.1 GPS Almanac............................................................................................... 39
A.10.2.2 GLONASS Almanac.................................................................................... 56
A.10.2.3 Galileo Almanac .......................................................................................... 69
A.10.2.4 BeiDou Almanac.......................................................................................... 83
A.10.2.5 SBAS Satellites............................................................................................ 96
A.10.3 Emulated errors ................................................................................................... 96

3
LIST OF FIGURES
Figure A-1: Chamber Mapping Grid ............................................................................................ 10
Figure A-2: Mapping Point differences between April 25th and April 28th................................ 12
Figure A-3: Histogram of Differences between April 25th and April 28th Mappings................. 13
Figure A-4: Power Differences across the Test Grid between April 25th and April 28th Mappings
(freqs. run from 1475 to 1675 at each point from left to right) .................................................... 14
Figure A-5: Mapping Point differences between April 22nd and April 25th............................... 15
Figure A-6: Histogram of Differences between April 22nd and April 25th Mappings................ 16
Figure A-7: Power Differences across the Test Grid between April 22nd and April 28th
Mappings (freqs. run from 1475 to 1675 at each point from left to right) ................................... 17
Figure A-8: Comparison of Mapping Corrections from DOT Test Week.................................... 18
Figure A-9: Comparison of Targeted Power (Recorded in Control File) and Measured Power
corrected for Spiral Antenna Gain and Cable Loss ...................................................................... 20
Figure A-10: Comparison of Targeted Power (Recorded in Control File) Differenced............... 21
Figure A-11: Diagram of Special Test where Transmit and Receive........................................... 22
Figure A-12: Measured Power at P33 (corrected for spiral gain and calibration cable) compared
with Predicted Power (interference system output power, horn antenna gain, horn cable loss, and
Free Space Loss)........................................................................................................................... 23
Figure A-13: AST-1507AA Positioning; Back View................................................................... 25
Figure A-14: AST-1507AA Boresight Measurements ................................................................. 26
Figure A-15: (a) A-Info Horizontal Position Side View, (b) A-Info Vertical Position Side Vie 27
Figure A-16: LB-510-10-C-NF Vertical Gain @ Boresight vs. Frequency ................................. 28
Figure A-17: LB-510-10-C-NF Vertical Pattern @ 1475, 1575 and 1675 MHz.......................... 29
Figure A-18: Spectra Showing Expanding Regrowth for Increasing HPA Input Power ............. 31
Figure A-19: Example Gain Response of Narrow and Relaxed Filters........................................ 35
Figure A-20: Composite PSDs for Innermost LTE Signals at -30 dBm....................................... 36
Figure A-21: Composite PSDs for Innermost LTE Signals at -10 dBm....................................... 37
Figure A-22: Composite PSDs for the Innermost Relaxed-filter LTE Signals at -10 dBm ......... 38
Figure A-23: GNSS Signal Generator Transmit Antenna Gain Patterns...................................... 39

4
LIST OF TABLES
Table A-1: Interference Transmitter Specifications ....................................................................... 6
Table A-2: LTE Parameters definition for the downlink and uplink signals.................................. 7
Table A-3: Example of Instrument settings for 1675 MHz.......................................................... 10
Table A-4: Calibration Values inserted into measlabtable.cfg:.................................................... 19
Table A-5: AST-1507AA, Serial #174......................................................................................... 26
Table A-6: Summary of WSMR Activities for April ABC Testing (4/4-4/28)............................ 29
Table A-7: Attenuation for Regrowth Spectra to Correspond to Radiated Tests ........................ 32

5
REFERENCES
There are no sources in the current document.

6
APPENDIX
WSMR Test Details
This Appendix provides a review of important aspects related to SPIGAT, the GNSS signal
generation, and activities performed during the DOT test week at WSMR. Included are
descriptions for generating the LTE signals, adjusting the mapping data to provide power
measurements across the equipment test area, analysis demonstrating measured vs. expected
power on the antenna floor, stability of power measurements over the test week, linearity of
power over test limits, system calibration, antenna and cable measurements used in these
analyses, and lastly, a summary of activities for the entire month at WSMR.
A.1 Interference System Development Considerations
The high level specifications for development of the SPIGAT capability are shown in Table A-1.
These specifications and the test configuration at the EMVAF drove several critical decisions in
development of SPIGAT. Three key issues were ensuring the maximum interference power
could be achieved across the entire test area, automating the interference tests as well as system
calibration and mapping, and lastly, ensuring interference signals were conditioned properly so
GNSS equipment degradation could be attributed to the fundamental signal being generated and
not unwanted emissions. Achieving the maximum power required a HPA that ultimately was
provided by the support engineers at WSMR. Additionally, satisfying maximum power also
required careful consideration of component insertion losses and ultimately sourcing a standard
gain horn designed for the interference frequency range to be tested. The desire to complete the
test effort in a single week dictated automation and sourcing an RF switch capable of handling
these power levels and frequencies to be tested. And lastly, the filters required performance
characteristics to sufficiently attenuate OOBE when generating LTE signals and led to
development of 22 RF passband cavity filters specifically for this test effort. Each of these areas
is touched on in later sections of this Appendix.
Table A-1: Interference Transmitter Specifications
Name Value Unit
Range of selectable center frequencies [1475, 1490, 1505, 1520, 1525, 1530,
1535, 1540, 1545, 1550, 1555, 1575,
1595, 1615, 1620, 1625, 1630, 1635,
1640, 1645, 1660, 1675]
MHz
Automation capability
Software controlled selection of center
frequency, signal type, and transmit
power
N/A
Interference signal generation
capability
White noise or LTE N/A

7
Number of selectable bandwidths 2 N/A
3 dB bandwidth (Signal Type-1 ) 1 MHz
3 dB bandwidth (Signal Type-2 ) Consistent with LTE N/A
Maximum power the setup is capable
of delivering at any receiver location
≥ -10 dBm
Another consideration for this testing was the LTE waveforms to be used for the downlink and
uplink signals. The LTE package offered by Keysight for its VSG products was considered
originally but ultimately it was determined the MATLAB emulation, which also adheres to the
LTE standard, had sufficient fidelity for this test effort. MATLAB generated time series
representation of the downlink and uplink signals and these were converted so they could be
loaded into the VSG for continuous replay. The actual MATLAB commands used to generate the
downlink and uplink waveforms are shown below in respective order.
[rmcwaveform_DL,rmcgrid_DL,rmccfgout_DL] =
lteRMCDLTool(rmcconfig_DL,rand([1,10000]));
[rmcwaveform_UL,rmcgrid_UL,rmccfgout_UL] =
lteRMCULTool(rmcconfig_UL,rand([1,10000]));
The associated configuration MATLAB structures used in the commands above are shown in
Table A-2. The duration of the uplink and downlink files was two seconds each and these were
played out continuously end to end from VSG memory.
Table A-2: LTE Parameters definition for the downlink and uplink signals
Downlink Parameters Definition Uplink Parameters Definition
Rmcconfig_DL.RC=‘R.2’ Rmcconfig_UL.RC = 'A3-5'
Rmcconfig_DL.NDLRB=50 Rmcconfig_UL.NULRB = 50
Rmcconfig_DL.CellRefP=1
Rmcconfig_DL.NCellID=0 Rmcconfig_UL.NCellID = 0
Rmcconfig_DL.CyclicPrefix=‘Normal’ Rmcconfig_UL.CyclicPrefixUL = 'Normal'
Rmcconfig_DL.CFI=2
Rmcconfig_DL.Ng=‘Sixth’
Rmcconfig_DL.PHICHDuration=‘Normal’
Rmcconfig_DL.NFrame=0 Rmcconfig_UL.NFrame = 0
Rmcconfig_DL.NSubframe=0 Rmcconfig_UL.NSubframe = 0
Rmcconfig_DL.TotSubframes=2000 Rmcconfig_UL.TotSubframes = 2000
Rmcconfig_UL.CyclicShift = 0
Rmcconfig_UL.Shortened = 0
Rmcconfig_UL.Hopping = 'Off'
Rmcconfig_DL.OCNG = ‘On’ Rmcconfig_UL.SeqGroup = 0

8
Rmcconfig_DL.Windowing = 16 Rmcconfig_UL.Windowing = 16
Rmcconfig_DL.DuplexMode = ‘FDD’ Rmcconfig_UL.DuplexMode = 'FDD'
Rmcconfig_UL.NTxAnts = 1
Rmcconfig_DL.SerialCat = 1
Rmcconfig_DL.SamplingRate = 15360000 Rmcconfig_UL.SamplingRate =15360000
Rmcconfig_DL.Nfft = 1024 Rmcconfig_UL.Nfft = 1024
Rmcconfig_DL.PDSCH.TxScheme = 'Port0'
Rmcconfig_DL.PDSCH.Modulation =
{'QPSK'}
Rmcconfig_UL.PUSCH.Modulation = 'QPSK'
Rmcconfig_DL.PDSCH.NLayers = 1 Rmcconfig_UL.PUSCH.NLayers =1
Rmcconfig_DL.PDSCH.Rho = 0
Rmcconfig_DL.PDSCH.RNTI = 1 Rmcconfig_UL.RNTI = 1
Rmcconfig_DL.PDSCH.RVSeq = [0 1 2 3] Rmcconfig_UL.PUSCH.RVSeq = [0 2 3 1]
Rmcconfig_DL.PDSCH.RV = 0 Rmcconfig_UL.PUSCH.RV = 0
Rmcconfig_DL.PDSCH.NHARQProcesses = 8 Rmcconfig_UL.PUSCH.NHARQProcesses =
8
Rmcconfig_DL.PDSCH.NTurboDecIts = 5 Rmcconfig_UL.PUSCH.NTurboDecIts = 5
Rmcconfig_UL.PUSCH.DynCyclicShift =0
Rmcconfig_UL.PUSCH.NBundled =0
Rmcconfig_UL.PUSCH.BetaACK =2
Rmcconfig_DL.PDSCH.CSIMode = 'PUCCH
1-1'
Rmcconfig_UL.PUSCH.BetaCQI = 2
Rmcconfig_DL.PDSCH.PMIMode =
'Wideband'
Rmcconfig_UL.PUSCH.BetaRI = 2
Rmcconfig_DL.PDSCH.PRBSet = [0:49] Rmcconfig_UL.PUSCH.PRBSet = [0:49]
Rmcconfig_DL.PDSCH.TrBlkSizes = [4392
4392 4392 4392 4392 0 4392 4392 4392 4392]
Rmcconfig_UL.PUSCH.TrBlkSizes = [5160
5160 5160 5160 5160 5160 5160 5160 5160
5160]
Rmcconfig_DL.PDSCH.CodedTrBlkSizes =
[12960 13800 13800 13800 13800 0 13800
13800 13800 13800]
Rmcconfig_UL.PUSCH.CodedTrBlkSizes =
[14400 14400 14400 14400 14400 14400
14400 14400 14400 14400]
Rmcconfig_UL.PUSCH.PMI = 0
Rmcconfig_UL.PUSCH.OrthCover = 'On'
A.2 Chamber Mapping

9
Chamber mapping determined RF power across the chamber floor and was ultimately used to
adjust SPIGAT test event summary files to establish power incident at each device tested.
Mapping was also accomplished with the GNSS signal generation system to demonstrate power
uniformity over the test area. The approach for mapping was to first establish a grid that
encompassed equipment placed in the test area. Figure A-1 shows the grid utilized in this effort
with 45 measurement points (blue diamonds) spaced approximately four feet apart. The
mapping grid origin for purposes of labeling points is at the bottom left corner of this figure
(point X=0 and Y=0). Each axis spanned values from 0 to 6. For reference, the mapping point
at the center of the grid was identified as P33 and was directly below the interference transmit
and GNSS antennas and P03 as will be described in a later section was the reference point used
for system calibrations. The red rectangles represent the area used for equipment under test.
The mapping and calibration procedure used a cavity backed spiral antenna mounted on a tripod,
calibrated RF cable, and a spectrum analyzer. The characteristics of the cavity backed spiral are
provided in a later section along with cable and transmit antenna characterizations. Power
measurements were made by positioning the tripod over a designated mapping point,
boresighting the cavity backed spiral on the transmit antenna using a laser pointer attached to the
antenna, having the interference generation and GNSS systems generate tones at each required
frequency, and lastly, having the spectrum analyzer step through each frequency and measure the
power of these tones. The interference signal generation system used the system calibration file
(described in a later section) to target a power level of -40 dBm as measured at the spiral
antenna. Measurements were taken with analyzer settings; auto-calibration disabled, preamplifier
ON with internal attenuation fixed at 6 dB (lowest available), span of 1 MHz using 1001 points
centered at the measurement frequency with resolution and video bandwidths of 5 kHz,
averaging ON and set to four frames, and using the RMS detector (see Table A-3).

10
Figure A-1: Chamber Mapping Grid
Table A-3: Example of Instrument settings for 1675 MHz
att: +6.00E+00 mkx: +1.6750000000000000E+009
autoatt: 0 mky: -5.38870E+01
autocal: OFF pdiv: +1.000E+01
avgstate: 1 peaky: 0
cont: 0 points: +1001
det: AVER preamp: 1
fcent: +1.6750000000000000E+009 rbw: +5.10000000E+003
fref: EXT reflev: -3.400E+01
fspan: +1.0000000000000000E+006 sweep: +1.46600000E-001
fstart: +1.6745000000000000E+009 trace: WRIT
fstop: +1.6755000000000000E+009 vavg: +4
vbw: +5.10000000E+003
A.3 Chamber Mapping Comparison and Stability

11
Chamber mapping at all 45 points was accomplished at the beginning (4/25) and end (4/28) of
DOT testing with all equipment installed in the test area. Mapping was also accomplished with
all equipment removed during the prior week (4/22) to establish variation introduced with GNSS
equipment in the test area. (Section A.8 provides a high level review of activities conducted for
the entire month of April at WSMR). For processing the mapping data, the power at each
frequency was adjusted for the target power of -40 dBm, spiral antenna gain at boresight and
calibration cable loss. Additionally, the calibration antenna was approximately 1.5 feet above
the blue absorbing material where GNSS equipment was placed for test. To account for this
difference, an additional 0.5 dB was included for this height differential to reflect the additional
free space loss. (Note the face of the transmit horn antenna was 25 feet above the blue absorbing
material so at these frequencies the difference of 1.5 feet equates to approximately 0.5 dB less
power received for equipment on the absorber compared with cavity backed spiral measured
power.) Ultimately, the corrected mapping measurements are interpolated to each equipment test
location and added to the interference system control file to determine the unique power received
at each of these locations.
The mappings performed on 4/25 and 4/28 were compared and found to be very consistent.
Figure A-2 shows a scatter plot of mapping location differences per frequency and demonstrates
the variation across these dates was within plus or minus approximately 0.5 dB. When the
differences are histogrammed by magnitude however (see Figure A-3), it can be observed that
over 98% of the mapping points agreed to within plus or minus 0.25 dB. Figure A-4 further
compares the power differences spatially and shows the lower left and upper right corners had
the largest variations. The final corrected mapping values provided for DOT receiver processing
represented the average of mapping from 4/25 and 4/28.
A mapping comparison was also made between 4/22 when no equipment was installed in the test
area and 4/25 to get a sense for how much the equipment affected received power. This can be
seen in Figure A-5 with a scatter plot of these differences. Notice the scale is double in this
figure compared with the previous scatter plot and differences were as large as plus or minus 1
dB. Figure A-6 shows the histogram of the differences and in this comparison approximately
75% of the mapping points agreed to within plus or minus 0.25 dB. The spatial comparison is
shown in Figure A-7. The largest differences are generally in the upper left corner which could
be due to concentration of equipment towards this end of the chamber. It should be noted that
increased variation between a clean chamber test area and one populated with over 80 pieces of
equipment under test is not surprising. This level of variation is actually considered very good
and was confirmed by the WSMR test engineers based on their experience with this chamber.

12
Figure A-2: Mapping Point differences between April 25th and April 28th

13
Figure A-3: Histogram of Differences between April 25th and April 28th Mappings

14
Figure A-4: Power Differences across the Test Grid between April 25th and April 28th
Mappings (freqs. run from 1475 to 1675 at each point from left to right)

15
Figure A-5: Mapping Point differences between April 22nd and April 25th

16
Figure A-6: Histogram of Differences between April 22nd and April 25th Mappings

17
Figure A-7: Power Differences across the Test Grid between April 22nd and April 28th
Mappings (freqs. run from 1475 to 1675 at each point from left to right)
One additional comparison noteworthy from mapping is how interference power varied over the
course of the DOT test week. The previous analysis demonstrated the variation for all 45
mapping locations at the start and end of testing. During the test week, a mapping procedure was
generally executed at reference location P03 in the AM and PM of each day as a confidence
check on test stability. Figure A-8 shows all P03 mappings during DOT week after being
corrected for target power, cavity backed spiral gain, calibration cable and the same 0.5 dB
mentioned above for height differential of the calibration antenna. The variation from morning
to evening and day to day was extremely well behaved with all measurements agreeing within
0.4 dB. For comparison purposes, the mapping at the reference location was also provided for
4/22 when no equipment was present in the test area. It is important to note that the calibration
antenna was moved numerous times over each day either for full mapping or to allow
participants access to the test area. Temperature was also recorded over the course of the one
month test effort and observed to vary by as much as 6 degrees C. This temperature variation
would be another source influencing daily power measurement variation.

18
Figure A-8: Comparison of Mapping Corrections from DOT Test Week
A.4 System Calibration
Calibration of the interference and GNSS systems were handled differently. This section
describes the interference system calibration only. The calibration procedure determined the
reference point on the grid as well as VSG settings to achieve desired power for each of the 22
interference test frequencies. The reference point for system calibration was determined by
finding the location receiving the least power from among grid points P03, P30, P36, and P63.
These peripheral points represent roughly the 3 dB beamwidth of the transmit antenna and
therefore selecting the location with the least received power ensures the majority of the test area
will receive the desired interference power or higher. Location P03 was determined as this point
and was used as the reference throughout all chamber testing. The calibration procedure was
very similar to mapping except it used a notional VSG setting as a starting point and then
computed power corrections per frequency to achieve -40 dBm at P03. This calculation used the
notional VSG setting, cavity backed spiral antenna gain, and calibration cable loss. During the
two week check out period prior to the formal test numerous system calibrations were executed.
Based on consistency of these measurements it was decided to take the linear average of these

19
values to fix the calibration table. The dates used in this average are provided below. Table A-4
provides the average calibration values.
cal_p03_20160408_144622, cal_p03_20160411_144033, cal_p03_20160412_142537,
cal_p03_20160413_131930, cal_p03_20160415_180323
Table A-4: Calibration Values inserted into measlabtable.cfg:
Frequency (MHz) Cal. Value (dB) Cal. Value (dB) Frequency (MHz)
1475 -3.30 1575 -4.90
1490 -3.21 1595 -5.29
1505 -3.34 1615 -6.08
1520 -3.69 1620 -6.03
1525 -3.50 1625 -6.27
1530 -3.61 1630 -6.11
1535 -3.59 1635 -6.01
1540 -3.70 1640 -6.22
1545 -3.78 1645 -6.52
1550 -4.09 1660 -6.53
1555 -4.28 1675 -6.71
One additional consideration for compensation of calibration values was use of a CW tone at
center frequency versus the LTE signal. The concern was the LTE signal would have additional
signal attenuation at filter edges due to roll-off vice the CW tone at center frequency. The signal
power loss for the LTE signal versus the single-tone was analyzed for each of the filters (6 and 8
section filters) and the maximum difference was 0.17 dB. Given this small value it was decided
adding a compensation term for LTE signal attenuation was not warranted.
A.5 Interference System Linearity
The interference linearity procedure measured interference power over the entire range for each
test frequency. This procedure was executed from P03 and stepped through the full range of
power in 2 dB steps with a CW tone. For the interference frequencies furthest from the RNSS
band the range was -80 to -10 dBm, for the three frequencies closet to and just above and below
the RNSS band the range was -100 to -30 dBm, and lastly, for the two frequencies in the RNSS
band the range was -130 to -60 dBm. As with calibration and mapping, the spiral antenna,
calibration cable and spectrum analyzer were used to measure and record each power
measurement. Figure A-9 shows measured power corrected for spiral antenna gain and cable loss

20
versus target power commanded by the interference system. This figure shows all frequency and
power measurements together and demonstrates the expected linear relationship from
approximately -100 dBm and higher. Below -100 dBm, the spectrum analyzer was not able to
accurately resolve the tone being received due to the measurement system noise floor. To further
demonstrate interference system performance, Figure A-10 shows the difference between
targeted and measured power for select frequencies. These figures show that over the power
range tested the interference system faithfully delivered intended power to within approximately
0.5 dB.
Figure A-9: Comparison of Targeted Power (Recorded in Control File) and Measured
Power corrected for Spiral Antenna Gain and Cable Loss

21
Figure A-10: Comparison of Targeted Power (Recorded in Control File) Differenced
from Measured Power for Select Frequencies
A.6 Comparison of Measured vs. Predicted Power
The measured power versus predicted power was estimated by making a special calibration of
the SPIGAT system. In this instance, interference system output was routed through a 60 dB
attenuator and into the spectrum analyzer used with the cavity backed spiral for making
measurements on the chamber floor. Therefore, cable/antenna to the transmit antenna and
cable/antenna from the receive antenna were bypassed (see Figure A-11) to obtain a direct power
measurement at the output of the interference system. The mapping procedure was executed as if
the calibration antenna was being utilized in the test area. The predicted power was then
computed using measured power from the spectrum analyzer corrected for the 60 dB attenuator
and associated cables, cable loss to the transmit antenna, transmit antenna gain, and free space
loss to the calibration antenna. This predicted power was then compared with measurements
made with the calibration antenna located at P33 which is directly below the transmit antenna
(peak beam).
FSL = -27.55 + 20 Log (fMHz) + 20 Log (Rmeter)

22
Rmeter = 7.16m (WSMR indicated face of horn to top of blue absorbing material directly
below is 25’. Calibration antenna is 1.5’ above absorbing material.)
Figure A-12 shows the result of the comparison using mapping data from 4/22 when the chamber
was clear of any test equipment. The figure demonstrates very good agreement between
predicted and measured power within approximately 0.5dB.
Figure A-11: Diagram of Special Test where Transmit and Receive
Antenna Paths were Bypassed

23
Figure A-12: Measured Power at P33 (corrected for spiral gain and calibration cable)
compared with Predicted Power (interference system output power, horn antenna gain,
horn cable loss, and Free Space Loss)
A.7 SPIGAT Antenna Characterizations
The Cavity Backed Spiral (Cobham AST-1507AA) and Standard Gain Horn (A-Info LB-510-10-
C-NF) antennas used in the test effort were provided to Leading Systems Technologies (LST) of
Fairfax, VA in March 2016 for gain and pattern characterizations. Characterization of the spiral
is key since it is required to calibrate received power of interference and GNSS signals across the
chamber test area while characterization of the horn is also desired to validate A-info’s product
data sheet and assist with link budget computations.
A.7.1 AST-1507AA

24
The critical estimates from the LST characterization in support of the ABC test effort are
measurements when the spiral is boresighted on the source. The orientation of the Cobham spiral
for testing at LST is provided in Figure A-13. The Figure represents the view from the back of
the antenna in the LST anechoic chamber and in this illustration the green dashed line and the
green arrow represent the axis of rotation for beam pattern observations. The beam pattern
measurements off boresight are not provided in this note but the 3 dB beamwidth was
determined to be approximately 80 degrees. For boresight characterization, the spiral antenna
was rotated sequentially from 0, 30, 45, 60, 90, 120, 135 and 150 degrees about the axis directed
toward the source. The orange rectangle represents the TNC female connector, which was
positioned facing up with respect to the ground as shown for the 0 degree orientation. The source
antenna was vertically polarized for these measurements due to better performance in this
chamber for the frequencies of interest. The frequency response of the spiral was measured from
800 to 2000 MHz inclusive of the frequencies of interest for ABC testing from 1227 to 1675
MHz.
Figure A-14 provides spiral antenna gain characteristics when boresighted at the source. The
similar response for each orientation of this antenna demonstrates its excellent axial ratio
properties. It is also interesting to note that variation with respect to antenna orientation above
1450 MHz is on the order of 0.25 dB which is reportedly approaching the accuracy limits for the
LST chamber. Table A-5 provides average gain and axial ratio versus frequencies required in the
ABC test effort. There are 22 frequencies utilized for interference generation and three specified
for GNSS antenna mapping. For purposes of completeness, the insertion loss of the RF cable
used at WSMR between the cavity backed spiral and the spectrum analyzer for mapping and
calibration acitivies has been included in this same table.

25
Figure A-13: AST-1507AA Positioning; Back View

26
Figure A-14: AST-1507AA Boresight Measurements
Table A-5: AST-1507AA, Serial #174
Frequency (MHz) Gain @ Boresight (dBil) Axial Ratio (dB) Cable Loss (dB)
1227* -0.98 0.50 7.76
1475. -0.24 0.20 8.44
1490. -0.11 0.10 8.42
1505. -0.18 0.10 8.45
1520. -0.28 0.10 8.52
1525. -0.34 0.10 8.54
1530. -0.24 0.10 8.57
1535. -0.25 0.10 8.60
1540. -0.35 0.10 8.63
1545. -0.29 0.20 8.64
1550. -0.16 0.20 8.66
1555. -0.07 0.30 8.72
1561* -0.06 0.30 8.74
1575. 0.21 0.30 8.76

27
1595. 0.17 0.30 8.78
1605* 0.45 0.40 8.76
1615. 0.44 0.30 8.75
1620. 0.41 0.30 8.75
1625. 0.65 0.30 8.77
1630. 0.73 0.30 8.79
1635. 0.56 0.30 8.83
1640. 0.54 0.20 8.87
1645. 0.69 0.20 8.88
1660. 0.70 0.20 9.00
1675. 0.86 0.30 9.08
*GNSS Antenna Mapping Frequencies for ABC Test
A.7.2 LB-510-10-C-NF
The A-Info horn antenna gain and pattern performance were characterized at two source antenna
orientations. For this testing, the source antenna and A-info polarizations were matched and the
horn was rotated to measure pattern response. Figure A-15(a) shows the A-Info position for
horizontal polarization measurements and (b) shows the position for vertical polarization
measurements. The frequency response of the horn was measured from 800 to 2000 MHz but the
frequencies of interest are primarily 1475 to 1675 MHz. Figure A-16 shows the horn gain at
boresight for vertical polarization and Figure A-17 shows beam pattern response for the same
polarization. Comparing these measurements with the A-Info product specification show a
reasonably good match.
Figure A-15: (a) A-Info Horizontal Position Side View,
(b) A-Info Vertical Position Side Vie
(a) (b)

28
Figure A-16: LB-510-10-C-NF Vertical Gain @ Boresight vs. Frequency

29
Figure A-17: LB-510-10-C-NF Vertical Pattern @ 1475, 1575 and 1675 MHz
A.8 WSMR Activity Summary
Table A-6 summarizes the activities for the entire month spent at WSMR. “F-Map” was a
complete 45 point mapping used in the final corrections of SPIGAT event summary files.
Table A-6: Summary of WSMR Activities for April ABC Testing (4/4-4/28)
Day AWGN LTE In-
band
IMD C/No 5-Map A-Map F-Map P03
Cal/
Map
Set-up/
Remove
M
T
W
Th
Arrived, unpacked equipment, installed transmit antennas, established grid, characterized HPA and cables,
dry-ran calibration and mapping, calibrated GNSS signals. Determined 1575 and 1595 MHz AWGN tests
needed separate runs due to inherent noise of HPA in-band.
Adjusted GNSS antenna lower at end of the 7th
to address reflections off horn (final location).
F
M2
4/081
4/11 4/11
4/08 4/08
4/11
4/08
4/11

30
T
W 4/13
4/12 4/12 4/12 4/12 4/12
4/13
4/12
4/13(2)
Th
F
M
T
W
Th
F
4/18
4/19 4/19
4/20
4/213
4/18
4/20
4/21
4/18
4/20
4/21
4/20
4/21
4/15
4/22
4/154
4/18(2)
4/19(2)
4/20
4/21
4/22
Set-up
Remove
M
T
W
Th
4/26
4/27
4/26
4/27
4/28 4/28
4/25
4/28
4/25
4/28
4/25
4/26
4/27
4/28
Set-up
Remove
1-“Eared” version of Type 2 LTE signal. All other Type-2 runs were with no “Ears”.
2-Characterization (system linearity) test run on 4/11.
3-No L2 signals.
4-Final calibration table established for all follow-on testing. Full mapping on 4/15 used this
table.
Checkout
DOD
Civil
A.9 Chamber OOBE
The fundamental components of the SPIGAT include a VSG to generate interference signals and
an HPA to provide sufficient amplification to achieve desired test levels. Given the limited
ACLR of a typical VSG and the potential for spectral regrowth due to HPA nonlinearity at high
input power, bandpass cavity filters were specified for SPIGAT with sufficient out of band
rejection to ensure that receiver CNR measurements would reflect degradation caused by the
fundamental of the interference, exclusive of any OOBE produced. The following test data and
analysis demonstrate that the OOBE suppression provided by SPIGAT during LTE radiated tests
performed at WSMR was sufficient to prevent receiver CNR degradation due to OOBE in the
RNSS band.
In SPIGAT the interference signal passes from the VSG through a programmable attenuator and
into the HPA. The amplified signal then passes through a bandpass cavity filter (specific to the
signal frequency and selected by RF switch) before being directed to devices under test. At

31
WSMR the signal was routed through a long cable to the transmit antenna where it was radiated
toward antennas on the chamber floor.
To assess the adequacy of OOBE suppression, spectra at the output of the HPA were recorded
via spectrum analyzer (connected through a 60 dB pad). Spectra were recorded for a range of
input power levels by fixing the signal power at the VSG while adjusting the programmable
attenuator in 2 dB increments over the desired range. These measurements were taken while
generating the downlink signal at 1550 MHz as this is the LTE test signal closest to the RNSS
band and, therefore, most relevant to the OOBE question. The expansion of spectral regrowth as
signal power increases is apparent from a suite of such measurements shown in Figure A-18.
Figure A-18: Spectra Showing Expanding Regrowth for Increasing HPA Input Power
The spectra thus recorded with the 60 dB pad are mapped to test levels by comparing VSG and
attenuator settings with those set during radiated tests. Table A-7 shows the correspondence.
Given VSG power fixed at 0 dBm (for regrowth spectra), the spectrum that corresponds to a

32
radiated test level is the one with attenuation nominally equal to the test attenuation less the test
VSG power. As an example, radiated tests of the 1550 MHz signal at -30 dBm are achieved with
VSG power of -5.91 dBm and 20 dB attenuation. Given VSG power fixed to 0 dBm, the
corresponding regrowth measurement has nominal attenuation of 26 dB.
Table A-7: Attenuation for Regrowth Spectra to Correspond to Radiated Tests
radiated tests regrowth measurements
A B C=A-B D E=D-C=B-A ~E
frequency
test
level
VSG
power Attenuation
attenuated
power
VSG
power
required
attenuation
nominal
attenuation
[MHz] [dBm] [dBm] [dB] [dBm] [dBm] [dB] [dB]
1550 -30 -5.91 20 -25.91 0 25.91 26
1620 -30 -3.97 20 -23.97 0 23.97 24
1550 -10 -5.91 0 -5.91 0 5.91 6
1620 -10 -3.97 0 -3.97 0 3.97 4
1535 -10 -6.41 0 -6.41 0 6.41 6
1630 -10 -3.89 0 -3.89 0 3.89 4
As noted above, regrowth spectra were recorded only for interference at 1550 MHz. In the
following, results presented for 1620 MHz (and other frequencies) were achieved by shifting
data taken at 1550 MHz. This seems appropriate because even though 1550 and 1620 MHz
interference signals were distinctly generated during radiated tests using the LTE downlink and
uplink signals, respectively, both signals produce a flat 9 MHz wide pedestal with steep roll off.
Given this similar energy distribution, the spectral regrowth for 1620 MHz interference is
expected to have similar shape. On the other hand, the HPA gain at 1620 MHz is about 0.5 dB
lower than at 1550 MHz. Therefore, regrowth at 1550 MHz may be slightly more severe than at
1620 MHz for a given input power since the higher gain at 1550 MHz moves the HPA farther
into compression. In summary, the 1550 MHz spectra applied to 1620 MHz may be slightly
pessimistic there (i.e. regrowth overstated).
For reference in the remaining discussion, the measured filter gain response of the 1550 and
1630 MHz filters are shown in Figure A-19 in the same scale as the figures that follow. The 1550
MHz filter is representative of the set of filters (including 1620 MHz) having narrow rolloff
constraints while the 1630 MHz filter is representative of the set having more relaxed
parameters.
Figure A-20 and Figure A-21 represent the 1550 and 1620 MHz LTE interference signals (blue
curves) as seen at the chamber floor for the -30 dBm test level (Figure A-20) and the -10 dBm
test level (Figure A-21). These two signals are selected because they are at the tested LTE

33
frequencies closest to and on either side of the RNSS band of interest. The curves were obtained
as follows. Spectra were selected with appropriate nominal attenuation for correspondence with
the target test level (see Table A-7). The measured filter response (1550 or 1620 MHz) was
added to the spectra to produce the composite post-filter signal shape. This approach of
measuring the HPA output and applying the filter response allows inspection of OOBE that may
otherwise be lost to the instrument noise floor if measured directly at the filter output, due to
steep rolloff of the cavity filters and limited dynamic range of the measuring device. Each
composite curve was then adjusted to the test level by subtracting the average of all points within
the 9 MHz pedestal and adding back the value of the test level (-30 or -10 dBm). Finally, the
curve was converted from units of dBm/9MHz to dBW/Hz by adding factors of -69.54 (to
convert /9 MHz to /Hz) and -30 (to convert dBm to dBW).
Regarding other figure elements, the vertical black lines mark the edges of the RNSS band. The
horizontal orange line is the ambient noise floor density, defined as 𝑁𝑁0 = 𝑘𝑘𝑘𝑘, for 𝑇𝑇 = 290 𝐾𝐾,
which is approximately -204 dBW/Hz. The size of additive OOBE relative to the ambient noise
floor determines the magnitude of C/N0 degradation experienced by a receiver. For example, the
horizontal yellow line, which is 16.3 dB lower than the orange line, is the level for additive noise
of constant density across the full receiver bandwidth which would produce 0.1 dB degradation
in the receiver. This is so merely because it is the noise density that when added to the ambient
noise density, raises the floor by 0.1 dB. OOBE is not likely to be of constant density, but the
line is included to serve as a useful reference. Receiver degradation is negligible (i.e less than 0.1
dB) for OOBE with density below the yellow line. Thus OOBE suppression can be assessed by
comparing the magnitude of OOBE with this reference.
Figure A-20 represents the interference signals at the -30 dBm test level using the 26 dB
regrowth spectra for 1550 MHz and the 24 dB spectra for 1620 MHz. This is the highest level
reached in radiated tests for these inner frequencies. By inspection of the plot the OOBE in the
RNSS band is below the yellow reference line, thereby causing negligible degradation. The
darker blue curve shows the portion of the regrowth spectra that was dominated by the noise
floor of the measuring instrument. From other measurements taken with better instrument
settings at higher HPA input power (less signal attenuation), it is known that the instrument noise
floor can be reduced by at least 12 dB without exposing any spectral regrowth at this power
level. For this reason a shadow noise floor (light grey) has been represented on the plot to
demonstrate that for the -30 dBm test level there is exceptional margin. This result also applies to
the other signals that are tested only to this level but farther from the RNSS band, namely the
signals at 1545 and 1625 MHz. In summary, OOBE suppression is more than adequate to
prevent receiver CNR degradation due to OOBE from the four LTE signals closest to the RNSS
band at their levels of test.
Figure A-21 represents the interference signals at the -10 dBm test level using the 6 dB regrowth
spectra for 1550 MHz and the 4 dB spectra for 1620 MHz. This level is above the test range for

34
these inner signals but was the original level used to specify filter attenuation and rolloff
performance to satisfy OOBE requirements. In this case, some OOBE within the RNSS band
rises above the yellow reference line, but it is in the extreme frequencies of the band, narrower
than 2 MHz between 1559 and 1561 MHz and narrower than 3 MHz between 1607 and 1610
MHz. The innermost signals tested at the -10 dBm level are at 1540 and 1630 MHz, each of
which is a full 10 MHz farther from the RNSS band. This additional separation is more than
enough for the 1540 MHz signal to shift the 2 MHz of OOBE to be outside the RNSS band, since
this signal uses a filter having the same narrow rolloff constraints as the 1550 and 1620 MHz
filters. However, application to the 1630 MHz signal is less apparent since this filter has more
relaxed parameters.
Figure A-22 represents the 1535 and 1630 MHz interference signals at the -10 dBm test level
using the 6 dB regrowth spectra for 1535 MHz and the 4 dB spectra for 1630 MHz. These two
frequencies are the innermost signals using filters having the more relaxed parameters. As
depicted, OOBE is well below the yellow 0.1 dB reference line, indicating receiver degradation
is negligible. This result also applies to the other signals that use filters having the relaxed
parameters, all of which are tested to this level but are farther from the RNSS band, namely 1475
through 1530 and 1635 through 1675 MHz.
In summary, the analysis demonstrates that OOBE suppression at WSMR was sufficient for all
levels of test at each LTE frequency. Figure A-20 demonstrates that OOBE is sufficiently
suppressed for all narrow-filter signals up through the -30 dBm test level (i.e. the full test range
for 1545, 1550, 1620 and 1625, but only up to -30 dBm for 1540); Figure A-21 demonstrates this
at the -10 dBm test level for all narrow-filter signals that run to this level (i.e. 1540; but also
1545 and 1625 although these are not actually tested at this level); and Figure A-22 demonstrates
this for all relaxed-filter signals up to the -10 dBm test level (i.e. the full test range for 1475
through 1535 and 1630 through 1675).

35
Figure A-19: Example Gain Response of Narrow and Relaxed Filters

36
Figure A-20: Composite PSDs for Innermost LTE Signals at -30 dBm

37
Figure A-21: Composite PSDs for Innermost LTE Signals at -10 dBm

38
Figure A-22: Composite PSDs for the Innermost Relaxed-filter LTE Signals at -10 dBm
A.10 GNSS Signal Generation
This section provides some details of the GNSS signal generation system.
A.10.1.1 Transmit antenna characterization
The gain patterns of the GNSS transmitting antenna at various center frequencies are shown in
Figure A-23. At boresight, this antenna provides RHCP gains of ~0.1 dBic (1227 MHz), 4.4 dBic
(1572 MHz), and 4.0 dBic (1602 MHz).

39
Figure A-23: GNSS Signal Generator Transmit Antenna Gain Patterns
A.10.2 Almanacs
A.10.2.1 GPS Almanac
0
345
330
315
300
285
270
255
240
225
210
195
180
165
150
135
120
105
90
75
60
45
30
15
-30 -20 -10 0
dB
Far-fieldam
plitudeofDsph16084a1.NSI
RHCP LHCP
(a) 1227.6 MHz
0
345
330
315
300
285
270
255
240
225
210
195
180
165
150
135
120
105
90
75
60
45
30
15
-30 -20 -10 0
dB
Far-fieldam
RHCP LHCP
(b) 1563 MHz
0
345
330
315
300
285
270
255
240
225
210
195
180
165
150
135
120
105
90
75
60
45
30
15
-30 -20 -10 0
dB
Far-fieldam
RHCP LHCP
(c) 1575 MHz
0
345
330
315
300
285
270
255
240
225
210
195
180
165
150
135
120
105
90
75
60
45
30
15
-30 -20 -10 0
dB
Far-fieldam
RHCP LHCP
(d) 1602 MHz

40
PRN-24 was generated at -10 dB and PRN-29 at -20 dB relative to the nominal received power
levels (see Table 7). Both of these satellites are visible throughout the entire 6 hour simulation
time. Using the GNSS generation equipment (see Section 4.3.3.1), due to a limitation on the
number of available channels the Spirent signal generator limits the number of emulated GPS
satellites to no more than ten based upon dilution of precision (DOP) (with PRN-24 and PRN-29
forced to always be present).
******** Week 869 almanac for PRN-01 ********
ID: 01
Health: 0
Eccentricity: 0.00000000000000e+000
Time of Applicability(s): 144000
Orbital Inclination(rad): 9.59931088700065e-001
Rate of Right Ascen(r/s): 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Right Ascen at Week(rad): 2.74046301800000e+000
Argument of Perigee(rad): -1.20197773000000e+000
Mean Anom(rad): -2.25293839000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 02
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

41
Argument of Perigee(rad): 2.71410071800000e+000
Mean Anom(rad): -1.21557569500000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 03
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Right Ascen at Week(rad): 4.49386715899998e-001
Argument of Perigee(rad): 8.89944077000001e-001
Mean Anom(rad): -1.57409238800000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 04
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

42
Mean Anom(rad): 1.62043118500000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 05
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Right Ascen at Week(rad): -5.81213951099999e-001
Mean Anom(rad): 1.22770595600000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 06
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

43
Mean Anom(rad): 9.50933694800001e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 07
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Right Ascen at Week(rad): -1.53364002700000e+000
Mean Anom(rad): -3.03474617000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 08
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

44
Mean Anom(rad): 2.57835054400000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 09
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 2.32180273500000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 10
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

45
Mean Anom(rad): -3.54773163799999e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 11
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 3.11031377300000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 12
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

46
Argument of Perigee(rad): -8.12766790000000e-001
Mean Anom(rad): -2.84381163100000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 13
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -5.35064935699999e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 14
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

47
Mean Anom(rad): -9.34581875800001e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 15
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -4.29327011100000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 16
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

48
Mean Anom(rad): 9.20198559800000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 17
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -1.65793895700000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 18
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

49
Mean Anom(rad): 1.04815793000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 19
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -9.07470583900000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 20
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

50
Mean Anom(rad): 1.19534623600000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 21
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 2.32440674300000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 22
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

51
Mean Anom(rad): -7.65956640200001e-002
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 23
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -1.44661092800000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 24
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

52
Mean Anom(rad): 1.03374876339972e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 25
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 1.46781361100000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 26
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

53
Mean Anom(rad): -1.88577055900000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 27
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 5.39627194399999e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 28
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

54
Mean Anom(rad): 2.58443236400000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 29
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): 2.46437074459972e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 30
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003

55
Mean Anom(rad): 6.90480113000000e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 31
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.15370081099980e+003
Mean Anom(rad): -3.10803759100000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869

56
A.10.2.2 GLONASS Almanac
SV-01 was generated at -10 dB and SV-02 at -20 dB relative to the nominal received power
time.
******** Week 869 almanac for SV-01, slot 1 ********
ID: 101
Health: 32
Eccentricity: 2.80435376435597e-004
Orbital Inclination(rad): 1.12033508367946e+000
SQRT(A) (m 1/2): 5.05032756696999e+003
N
E
S
W
2
4
5
9
10
12 17
24
28
29
30
3
6
14
15
16
18
21
22
24
26
29

57
Mean Anom(rad): 1.15091734896345e+000
Af0(s): -1.45391561091000e-004
Af1(s/s): 0.00000000000000e+000
week: 869
******** Week 869 almanac for SV-02, slot -4 ********
ID: 2-4
Health: 32
Eccentricity: 1.33246975536473e-003
SQRT(A) (m 1/2): 5.05064077398080e+003
Mean Anom(rad): -2.59971956740604e+000
Af0(s): 1.79462134838000e-004
Af1(s/s): 1.81898940355000e-012
week: 869
ID: 305
Health: 32
Eccentricity: 1.32498816166167e-003
SQRT(A) (m 1/2): 5.05066930031813e+003

58
Mean Anom(rad): 1.69623491824316e+000
Af0(s): 6.39893114567000e-005
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 406
Health: 32
Eccentricity: 4.07477957999495e-004
SQRT(A) (m 1/2): 5.05036637505249e+003
Mean Anom(rad): 1.15466300306374e+000
Af0(s): 1.59780494869000e-004
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 501
Health: 32
Eccentricity: 5.25326160780691e-004
SQRT(A) (m 1/2): 5.05034296657956e+003

59
Mean Anom(rad): 3.13126501826303e+000
Af0(s): -3.61446291208000e-006
Af1(s/s): -9.09494701773000e-013
week: 869
ID: 6-4
Health: 32
Eccentricity: 8.53360015146096e-004
SQRT(A) (m 1/2): 5.05065643919592e+003
Mean Anom(rad): 1.08577069852623e+000
Af0(s): 8.49086791277000e-005
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 705
Health: 32
Eccentricity: 1.10069518293223e-003
SQRT(A) (m 1/2): 5.05066586996986e+003

60
Mean Anom(rad): -1.40932465105142e+000
Af0(s): -6.72508031130000e-006
Af1(s/s): -0.00000000000000e+000
week: 869
ID: 806
Health: 32
Eccentricity: 1.69977794661984e-003
SQRT(A) (m 1/2): 5.05037218492650e+003
Mean Anom(rad): -2.86620078192880e+000
Af0(s): -2.71061435342000e-005
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 9-6
Health: 32
Eccentricity: 1.49164382262733e-003
SQRT(A) (m 1/2): 5.05029813945331e+003

61
Mean Anom(rad): -2.65756649912442e+000
Af0(s): 8.67033377290000e-005
Af1(s/s): 7.27595761418000e-012
week: 869
ID: 10-7
Health: 32
Eccentricity: 1.64202181019619e-003
SQRT(A) (m 1/2): 5.05053138049315e+003
Mean Anom(rad): -1.96468327760479e+000
Af0(s): 1.51759013534000e-005
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 1100
Health: 32
Eccentricity: 1.96070028249679e-003
SQRT(A) (m 1/2): 5.05072659215691e+003

62
Mean Anom(rad): -1.57173974293076e-001
Af0(s): 2.48514115810000e-005
Af1(s/s): -0.00000000000000e+000
week: 869
ID: 12-4
Health: 32
Eccentricity: 3.29010000000000e-003
SQRT(A) (m 1/2): 5.05054067518879e+003
Mean Anom(rad): 1.79044601315838e-001
Af0(s): 5.00000000000000e-004
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 13-2
Health: 32
Eccentricity: 7.99396667097814e-004
SQRT(A) (m 1/2): 5.05028282604365e+003

63
Mean Anom(rad): 3.13950706608156e+000
Af0(s): 1.00303441286000e-005
Af1(s/s): -0.00000000000000e+000
week: 869
ID: 14-7
Health: 32
Eccentricity: 1.55466745629980e-003
SQRT(A) (m 1/2): 5.05056499236033e+003
Mean Anom(rad): 1.14646984845523e+000
Af0(s): 2.89653427899000e-004
Af1(s/s): 4.54747350886000e-012
week: 869
ID: 1500
Health: 32
Eccentricity: 2.23977675463663e-003
SQRT(A) (m 1/2): 5.05071089313078e+003

64
Mean Anom(rad): -3.08905125954067e+000
Af0(s): 6.49411231279000e-005
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 16-1
Health: 32
Eccentricity: 1.83550000000000e-003
SQRT(A) (m 1/2): 5.05054371669493e+003
Mean Anom(rad): 6.91782192978976e-001
Af0(s): 5.00000000000000e-004
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 1704
Health: 32
Eccentricity: 8.84576720774307e-004
SQRT(A) (m 1/2): 5.05029194768975e+003

65
Mean Anom(rad): -2.16435670581638e+000
Af0(s): -2.28835269809000e-005
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 18-3
Health: 32
Eccentricity: 1.58283047580818e-003
SQRT(A) (m 1/2): 5.05041473916712e+003
Mean Anom(rad): 1.47504581863733e+000
Af0(s): 9.16374847293000e-005
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 1903
Health: 32
Eccentricity: 2.44352903735194e-004
SQRT(A) (m 1/2): 5.05068914526964e+003

66
Mean Anom(rad): 1.36025505659405e+000
Af0(s): -1.49535015225000e-004
Af1(s/s): -0.00000000000000e+000
week: 869
ID: 2002
Health: 32
Eccentricity: 1.22842522658171e-003
SQRT(A) (m 1/2): 5.05054064557828e+003
Mean Anom(rad): -2.46146248236104e-001
Af0(s): -2.22451053560000e-004
Af1(s/s): -1.81898940355000e-012
week: 869
ID: 2104
Health: 32
Eccentricity: 5.31475615345574e-004
SQRT(A) (m 1/2): 5.05029468247230e+003

67
Mean Anom(rad): 1.44889418884053e+000
Af0(s): 5.36972656846000e-005
Af1(s/s): 1.81898940355000e-012
week: 869
ID: 22-3
Health: 32
Eccentricity: 2.94760840310346e-003
SQRT(A) (m 1/2): 5.05043482102626e+003
Mean Anom(rad): -2.05717394866684e+000
Af0(s): 1.93633139133000e-004
Af1(s/s): 9.09494701773000e-013
week: 869
ID: 2303
Health: 32
Eccentricity: 3.94066120249689e-005
SQRT(A) (m 1/2): 5.05071557599688e+003

68
Mean Anom(rad): -2.81819268437635e+000
Af0(s): -8.79149883986000e-005
Af1(s/s): -3.63797880709000e-012
week: 869
ID: 2402
Health: 32
Eccentricity: 6.68632423028246e-004
SQRT(A) (m 1/2): 5.05057594902682e+003
Mean Anom(rad): 1.43907705778873e+000
Af0(s): -1.34304165840000e-004
Af1(s/s): 0.00000000000000e+000
week: 869

69
A.10.2.3 Galileo Almanac
time.
ID: 01
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
N
E
S
W
1
2
7
8
9
16
21
22
23
1
2
3
11
12
13
14

70
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 02
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 6.98131700797731e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 03
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

71
Mean Anom(rad): 1.39626340159546e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 04
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 2.09439510239320e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 05
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

72
Mean Anom(rad): 2.79252680319093e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 06
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -2.79252680319093e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 07
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

73
Mean Anom(rad): -2.09439510239320e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 08
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -1.39626340159547e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 09
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

74
Mean Anom(rad): -6.98131700797731e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 10
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 2.32652389290844e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 11
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

75
Mean Anom(rad): 9.30784090088577e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 12
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 1.62891579088631e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 13
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

76
Mean Anom(rad): 2.32704749168404e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 14
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 3.02517919248177e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 15
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

77
Mean Anom(rad): -2.55987441390008e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 16
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -1.86174271310235e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 17
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

78
Mean Anom(rad): -1.16361101230462e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 18
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -4.65479311506889e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 19
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

79
Mean Anom(rad): 4.65304778581688e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 20
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 1.16343647937942e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 21
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

80
Mean Anom(rad): 1.86156818017715e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 22
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): 2.55969988097488e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 23
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

81
Mean Anom(rad): -3.02535372540697e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 24
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -2.32722202460924e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 25
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

82
Mean Anom(rad): -1.62909032381151e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 26
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003
Mean Anom(rad): -9.30958623013774e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 27
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.44058820349418e+003

83
Mean Anom(rad): -2.32826922216043e-001
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
A.10.2.4 BeiDou Almanac
N
E
S
W
3
4
5
6
10
11
12
13
23
24
25
1
2
3
4
10
17
18
24
25
26

84
time.
ID: 06
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 07
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

85
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 08
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 09
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

86
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 10
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 11
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

87
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 12
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 13
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

88
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 14
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 15
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

89
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 16
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 17
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

90
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 18
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 19
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

91
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 20
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 21
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

92
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 22
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 23
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

93
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 24
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 25
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

94
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 26
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 27
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

95
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 28
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
ID: 29
Health: 0
Eccentricity: 0.00000000000000e+000
SQRT(A) (m 1/2): 5.27996000000000e+003

96
Mean Anom(rad): 0.00000000000000e+000
Af0(s): 0.00000000000000e+000
Af1(s/s): 0.00000000000000e+000
week: 869
A.10.2.5 SBAS Satellites
The two SBAS satellites were emulated in the geostationary satellite arc above 133W and 108W
with PRNs 135 and 138, respectively.
A.10.3 Emulated errors
N
E
S
W
8 9
10
11
18
19
20
21
22
23
29
6
7
8
16
17
18
22
23
24

97
The Spirent RTCA06 tropospheric model was utilized to generate tropospheric errors on the
pseudorange and carrier phase measurements as appropriate for the emulated user location and
date (32N, 106W, April 18, 2016).
Ionospheric errors were emulated using the GPS single-frequency ionospheric delay model (i.e.,
the Klobuchar model). The following 8 coefficients were utilized:
ALPHA0: 5.58793544769E-009
ALPHA1: 1.49011611938E-008
ALPHA2: -5.96046447754E-008
ALPHA3: -1.19209289551E-007
BETA0: 83968
BETA1: 98304
BETA2: -65536
BETA3: -524288
Galileo uses a different single-frequency ionospheric correction model known as NeQuick.
NeQuick coefficients were generated, using a Spirent GSS8000 simulator software utility, to best
match ionospheric delay corrections per the GPS Klobuchar model with the coefficients listed
above.

1
APPENDIX B
AGGREGATE ITM RESULTS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 6
REFERENCES ............................................................................................................................... 7
APPENDIX..................................................................................................................................... 8
Aggregate ITM Results........................................................................................... 8
B.1 Statistical and Bounding ITMs for 1MHz and 10 MHz LTE Interference Signals ....... 8
B.1.1 Interference Test Results for 1-MHz AWGN ......................................................... 8
B.1.2 Interference Test Results for 10-MHz LTE........................................................... 31
B.1.3 Combined 1 MHz AWGN Test Results and 10 MHz LTE Test Results .............. 55

3
LIST OF FIGURES
Figure B-1: 1 MHz GPS L1 C/A Statistical Mask Results for Cellular receivers......................... 8
Figure B-2: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for Cellular
receivers .......................................................................................................................................... 9
Figure B-3: 1 MHz GPS L1 C/A Statistical Mask Results for General Aviation receivers ........ 10
Figure B-4: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for General
Aviation receivers ......................................................................................................................... 10
Figure B-5: 1 MHz GPS L1 C/A Statistical Mask Results for General Location Navigation
receivers ........................................................................................................................................ 11
Location Navigation receivers ...................................................................................................... 12
Figure B-7: 1 MHz GPS L1 C/A Statistical Mask Results for High Precision receivers............ 12
Figure B-8: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for High Precision
receivers ........................................................................................................................................ 13
Figure B-9: 1 MHz GPS L1 C/A Statistical Mask Results for Space Based receivers ............... 14
Figure B-10: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for Space Based
receivers ........................................................................................................................................ 14
Figure B-11: 1 MHz GPS L1 C/A Statistical Mask Results for Timing receivers...................... 15
Figure B-12: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for Timing
receivers ........................................................................................................................................ 16
Figure B-13: Summary of 1 MHz and 1 MHz In-band GPS L1 C/A Bounding Masks............... 17
Figure B-14: Summary of 1 MHz and 1 MHz In-band GPS L1 P Bounding Masks ................... 19
Figure B-15: Summary of 1 MHz and 1 MHz In-band GPS L1 C Bounding Masks................... 21
Figure B-16: Summary of 1 MHz and 1 MHz In-band GLONASS L1 C Bounding Masks........ 23
Figure B-17: Summary of 1 MHz and 1 MHz In-band GLONASS L1 P Bounding Masks ........ 25
Figure B-18: Summary of 1 MHz and 1 MHz In-band BeiDou B1 I Bounding Masks............... 27
Figure B-19: Summary of 1 MHz and 1 MHz In-band Galileo E1 BC Bounding Masks............ 29
Figure B-20: Summary of 1 MHz and 1 MHz In-band SBAS L1 C/A Bounding Masks ............ 31
Figure B-21: 10 MHz GPS L1 C/A Statistical Mask Results for Cellular receivers................... 32
Figure B-22: 10 MHz All GNSS Aggregated Minimum Statistical Mask Results for Cellular
receivers ........................................................................................................................................ 33
Figure B-23: 10 MHz GPS L1 C/A Statistical Mask Results for General Aviation receivers .... 33
Aviation receivers ......................................................................................................................... 34
Figure B-25: 10 MHz GPS L1 C/A Statistical Mask Results for General Location Navigation
receivers ........................................................................................................................................ 35

4
Location Navigation receivers ...................................................................................................... 35
Figure B-27: 10 MHz GPS L1 C/A Statistical Mask Results for High Precision receivers........ 36
Figure B-28: 10 MHz All GNSS Aggregated Minimum Statistical Mask Results for High
Precision receivers ........................................................................................................................ 37
Figure B-29: 10 MHz GPS L1 C/A Statistical Mask Results for Space Based receivers ........... 37
Figure B-30: 10 MHz All GNSS Aggregated Minimum Statistical Mask Results for Space
Based receivers ............................................................................................................................. 38
Figure B-31: 10 MHz GPS L1 C/A Statistical Mask Results for Timing receivers.................... 39
Figure B-32: 10 MHz All GNSS Aggregated Minimum Statistical Mask Results for Timing
receivers ........................................................................................................................................ 39
Figure B-33: Summary of 10 MHz and 10MHz Intermod GPS L1 C/A Bounding Masks.......... 41
Figure B-34: Summary of 10 MHz and 10 MHz Intermod GPS L1 P Bounding Masks............. 43
Figure B-35: Summary of 10 MHz and 10 MHz Intermod GPS L1 C Bounding Masks............. 45
Figure B-36: Summary of 10 MHz and 10 MHz Intermod GLONASS L1 C Bounding Masks . 47
Figure B-37: Summary of 10 MHz and 10 MHz Intermod GLONASS L1 P Bounding Masks.. 49
Figure B-38: Summary of 10 MHz and 10 MHz Intermod BeiDou B1 I Bounding Masks......... 51
Figure B-39: Summary of 10 MHz and 10 MHz Intermod Galileo E1 BC Bounding Masks ..... 53
Figure B-40: Summary of 10 MHz and 10 MHz Intermod SBAS L1 C/A Bounding Masks...... 55
Figure B-41: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GPS L1 C/A –
HPR Bounding Masks................................................................................................................... 56
Figure B-42: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod with Certified
Aviation GPS L1 C/A Bounding Masks....................................................................................... 57
Figure B-43: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GPS L1 P – HPR
Bounding Masks ........................................................................................................................... 58
Figure B-44: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GPS L1 P
Bounding Masks ........................................................................................................................... 59
Figure B-45: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GPS L1 C - GLN
Bounding Masks ........................................................................................................................... 60
Figure B-46: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GPS L1 C
Bounding Masks ........................................................................................................................... 61
Figure B-47: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GLONASS L1 C
- HPR Bounding Masks ................................................................................................................ 62
Figure B-48: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GLONASS L1 C
Bounding Masks ........................................................................................................................... 63
Figure B-49: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GLONASS L1 P
- HPR Bounding Masks ................................................................................................................ 64
Figure B-50: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod GLONASS L1 P
Bounding Masks ........................................................................................................................... 65

5
Figure B-51: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod BeiDou B1 I -
Figure B-52: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod BeiDou B1 I
Bounding Masks ........................................................................................................................... 67
Figure B-53: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod Galileo E1 BC
Bounding Masks ........................................................................................................................... 68
Figure B-54: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod SBAS L1 C/A -
Figure B-55: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod SBAS L1 C/A
Bounding Masks ........................................................................................................................... 70
Figure B-56: Summary of Cellular Bounding Masks for 1 & 10 MHz, 1 MHz In-band, and 10
MHz Intermod............................................................................................................................... 71
Figure B-57: Summary of General Aviation Bounding Masks for 1 & 10 MHz, 1 MHz In-band,
and 10 MHz Intermod................................................................................................................... 72
Figure B-58: Summary of General Location Navigation Bounding Masks for 1 & 10 MHz, 1
MHz In-band, and 10 MHz Intermod ........................................................................................... 73
Figure B-59: Summary of High Precision Bounding Masks for 1 & 10 MHz, 1 MHz In-band,
and 10 MHz Intermod................................................................................................................... 74
Figure B-60: Summary of Space Based Bounding Masks for 1 & 10 MHz, 1 MHz In-band, and
10 MHz Intermod.......................................................................................................................... 75
Figure B-61: Summary of Timing Bounding Masks for 1 & 10 MHz, 1 MHz In-band, and 10
MHz Intermod............................................................................................................................... 76

6
LIST OF TABLES
Table B-1: 1 MHz and 1 MHz In-band* GPS L1 C/A Bounding Masks (dBm).......................... 16
Table B-2: 1 MHz and 1 MHz In-band* GPS L1 P Bounding Masks (dBm).............................. 18
Table B-3: 1 MHz and 1 MHz In-band* GPS L1 C Bounding Masks (dBm).............................. 20
Table B-4: 1 MHz and 1 MHz In-band* GLONASS L1 C Bounding Masks (dBm) .................. 22
Table B-5: 1 MHz and 1 MHz In-band* GLONASS L1 P Bounding Masks (dBm)................... 24
Table B-6: 1 MHz and 1 MHz In-band* BeiDou B1 I Bounding Masks (dBm).......................... 26
Table B-7: 1 MHz and 1 MHz In-band* Galileo E1 BC Bounding Masks (dBm) ...................... 28
Table B-8: 1 MHz and 1 MHz In-band* SBAS L1 C/A Bounding Masks (dBm)....................... 30
Table B-9: 10 MHz and 10 MHz Intermod* GPS L1 C/A Bounding Masks (dBm) ................... 40
Table B-10: 10 MHz and 10 MHz Intermod* GPS L1 P Bounding Masks (dBm)...................... 42
Table B-11: 10 MHz and 10 MHz Intermod* GPS L1 C Bounding Masks (dBm) ..................... 44
Table B-12: 10 MHz and 10 MHz Intermod* GLONASS L1 C Bounding Masks (dBm) .......... 46
Table B-13: 10 MHz and 10 MHz Intermod* GLONASS L1 P Bounding Masks (dBm)........... 48
Table B-14: 10 MHz and 10 MHz Intermod* BeiDou B1 I Bounding Masks (dBm) ................. 50
Table B-15: 10 MHz and 10 MHz Intermod* Galileo E1 BC Bounding Masks (dBm) .............. 52
Table B-16: 10 MHz and 10 MHz Intermod* SBAS L1 C/A Bounding Masks (dBm)............... 54

7
REFERENCES

8
APPENDIX
Aggregate ITM Results
B.1 Statistical and Bounding ITMs for 1MHz and 10 MHz LTE Interference Signals
B.1.1 Interference Test Results for 1-MHz AWGN
Figure B-1: 1 MHz GPS L1 C/A Statistical Mask Results for
Cellular receivers

9
Figure B-2: 1 MHz All GNSS Aggregated Minimum Statistical Mask Results for
Cellular receivers

10
General Aviation receivers

11
General Location Navigation receivers

12
High Precision receivers

13

14
Space Based receivers

15
Timing receivers

16
Timing receivers
Table B-1: 1 MHz and 1 MHz In-band* GPS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
GAV GLN HPR TIM SPB CEL
1475 -35.2559 -51.0577 -55.2458 -14.2455 -44.2989 -12.0223
1490 -33.7621 -57.3312 -60.0236 -16.8303 -54.552 -12.6553
1505 -29.8401 -43.0491 -60.4615 -23.7082 -68.4079 -13.2884
1520 -34.9844 -51.2192 -68.9024 -34.9844 -75.2401 -13.9214
1525 -43.1501 -53.1671 -76.7213 -43.1501 -75.2341 -14.1324
1530 -48.1543 -52.0402 -68.0809 -48.1543 -74.5091 -14.6318
1535 -58.124 -52.322 -72.0429 -58.124 -69.3135 -15.1311
1540 -62.9739 -52.9021 -77.1205 -62.9739 -74.6895 -30.9017
1545 -72.8509 -61.3673 -91.2872 -72.8509 -79.5006 -41.9222
1550 -82.2979 -75.7539 -92.663 -82.2979 -82.1933 -50.8924
1555 -87.2361 -83.7667 -92.0068 -93.6384 -80.8092 -54.9029
1575* -119.066 -119.066 -116.191 -115.066 -110.143 -113.425
1595* -82.7539 -83.6945 -99.8672 -82.7539 -81.6271 -75.4419
1615 -48.8394 -73.2344 -91.4798 -69.7614 -77.6033 -65.0079
1620 -42.5903 -66.3442 -80.6362 -48.8434 -77.2055 -54.0767
1625 -52.9903 -54.7378 -65.5759 -36.1468 -84.4141 -42.0742
1630 -37.111 -55.4588 -63.0348 -29.2349 -74.4229 -23.7846
1635 -33.9518 -43.9075 -65.1038 -22.3593 -72.5191 -13.8657
1640 -39.2495 -43.7996 -60.2043 -18.0435 -74.405 -14.2567
1645 -27.2708 -43.2334 -62.1873 -17.0656 -74.1188 -13.18
1660 -31.3256 -42.1121 -55.9123 -20.8014 -71.7159 -14.0507
1675 -43.4325 -34.4705 -44.4567 -18.0372 -65.8974 -13.9362

17
Figure B-13: Summary of 1 MHz and 1 MHz In-band GPS L1 C/A Bounding Masks

18
Table B-2: 1 MHz and 1 MHz In-band* GPS L1 P Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR SPB CEL
1475 -9.999 -54.6169 -16.4489 -12.0223
1490 -9.999 -60.4599 -57.1999 -12.6553
1505 -9.999 -62.3551 -57.8913 -13.2884
1520 -9.999 -65.2899 -58.5827 -13.9214
1525 -9.999 -62.0257 -58.8132 -14.1324
1530 -9.999 -62.0352 -59.7591 -14.6318
1535 -10.5239 -62.5546 -75.151 -15.1311
1540 -13.6638 -62.5322 -78.6008 -19.999
1545 -29.999 -62.6246 -82.0506 -29.999
1550 -40.4331 -64.2835 -78.9579 -29.999
1555 -50.8672 -66.7664 -83.1592 -29.999
1575* -111.658 -108.025 -112.731 -113.227
1595* -70.3132 -66.6468 NaN -75.4419
1615 -66.3527 -63.5333 -81.9033 -65.0079
1620 -58.9925 -62.7395 -79.9751 -50.6889
1625 -43.759 -62.5091 -87.4391 -35.3214
1630 -30.4904 -61.0547 -84.1715 -23.7846
1635 -27.0752 -61.7723 -80.9039 -13.8657
1640 -23.66 -61.6717 -77.6363 -14.2567
1645 -20.2447 -62.4248 -74.3688 -13.18
1660 -9.999 -50.8114 -74.2659 -14.0507
1675 -9.999 -40.097 -65.4474 -13.9362

19
Figure B-14: Summary of 1 MHz and 1 MHz In-band GPS L1 P Bounding Masks

20
Table B-3: 1 MHz and 1 MHz In-band* GPS L1 C Bounding Masks (dBm)
Frequency
(MHz)
GLN CEL
1475 -9.999 -12.0223
1490 -9.999 -12.6553
1505 -9.999 -13.2884
1520 -9.999 -13.9214
1525 -9.999 -14.1324
1530 -9.999 -14.6318
1535 -10.5239 -15.1311
1540 -13.6638 -19.999
1545 -29.999 -29.999
1550 -33.2037 -29.999
1555 -50.8672 -29.999
1575* -111.658 -113.227
1595* -70.3132 -75.4419
1615 -66.3527 -65.0079
1620 -58.9925 -50.6889
1625 -43.759 -35.3214
1630 -30.4904 -23.7846
1635 -27.0752 -13.8657
1640 -23.66 -14.2567
1645 -20.2447 -13.18
1660 -9.999 -14.0507
1675 -9.999 -13.9362

21
Figure B-15: Summary of 1 MHz and 1 MHz In-band GPS L1 C Bounding Masks

22
Table B-4: 1 MHz and 1 MHz In-band* GLONASS L1 C Bounding Masks (dBm)
Frequency
(MHz)
GAV GLN HPR TIM CEL
1475 -38.2559 -44.7577 -54.6169 -15.2559 -15.4348
1490 -53.4803 -53.4803 -62.9099 -19.5303 -14.0758
1505 -48.4582 -48.4582 -64.9422 -25.7082 -12.7169
1520 -56.1318 -56.1318 -76.9467 -31.3318 -11.3579
1525 -59.3929 -59.3929 -76.8713 -35.5429 -10.905
1530 -61.4793 -61.4793 -76.7309 -36.8043 -10.452
1535 -63.145 -63.145 -76.5364 -38.695 -9.999
1540 -65.4252 -65.4252 -76.3935 -49.3752 -21.2267
1545 -69.3673 -69.3673 -79.0175 -71.1694 -38.8222
1550 -72.6693 -72.6693 -86.7862 -72.6693 -42.0942
1555 -77.2525 -77.2525 -92.9286 -90.0964 -50.9955
1575* -74.1159 -77.5833 -106.008 -90.6933 -73.3266
1595* -98.5561 -101.975 -113.167 -110.561 -79.5011
1615 -49.3102 -75.1443 -96.917 -75.1443 -66.5954
1620 -39.5942 -64.2107 -90.419 -64.2107 -47.9014
1625 -43.7848 -52.0968 -80.2284 -52.0968 -37.3339
1630 -30.8828 -44.72 -70.3042 -39.5099 -25.5447
1635 -29.1745 -42.9075 -69.3723 -28.8593 -17.7719
1640 -25.6797 -44.4996 -70.2717 -26.931 -15.5567
1645 -18.4429 -37.7709 -68.1748 -23.0656 -16.18
1660 -15.3092 -40.1121 -61.9614 -22.0264 -14.3642
1675 -9.999 -37.4705 -55.797 -22.5372 -14.3862

23
Figure B-16: Summary of 1 MHz and 1 MHz In-band GLONASS L1 C Bounding Masks

24
Table B-5: 1 MHz and 1 MHz In-band* GLONASS L1 P Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR CEL
1475 -9.999 -9.999 -15.4348
1490 -9.999 -9.999 -14.0758
1505 -9.999 -11.8353 -12.7169
1520 -12.0311 -13.6716 -11.3579
1525 -9.999 -18.1971 -10.905
1530 -9.999 -26.0448 -10.452
1535 -9.999 -27.7146 -9.999
1540 -12.7888 -30.2492 -19.999
1545 -29.999 -38.772 -29.999
1550 -37.5179 -51.3101 -29.999
1555 -50.9922 -58.7847 -29.999
1575* -77.5833 -100.344 -73.3266
1595* -101.975 -105.436 -75.3544
1615 -64.1152 -91.7591 -66.5954
1620 -59.0175 -76.0987 -47.9014
1625 -45.8712 -48.0537 -37.3339
1630 -33.3779 -30.6987 -21.0846
1635 -27.6843 -24.6631 -16.1157
1640 -12.8654 -20.8626 -15.5567
1645 -12.1488 -19.2505 -16.18
1660 -9.999 -14.6248 -14.3642
1675 -9.999 -9.999 -14.3862

25
Figure B-17: Summary of 1 MHz and 1 MHz In-band GLONASS L1 P Bounding Masks

26
Table B-6: 1 MHz and 1 MHz In-band* BeiDou B1 I Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR CEL
1475 -15.2553 -9.999 -9.999
1490 -18.5174 -26.9935 -9.999
1505 -22.5464 -49.829 -9.999
1520 -28.4477 -63.8777 -9.999
1525 -29.3591 -71.0896 -9.999
1530 -34.3999 -76.0923 -9.999
1535 -35.2722 -76.6179 -14.7446
1540 -36.3339 -77.9462 -19.4902
1545 -61.2885 -87.3658 -42.9971
1550 -71.2057 -94.8011 -65.9259
1555 -90.2237 -96.4363 -74.9061
1575* -78.1883 -100.556 -65.7427
1595* -77.9432 -93.2012 NaN
1615 -80.688 -79.3254 -33.2232
1620 -53.752 -76.038 -29.999
1625 -37.6608 -68.782 -29.999
1630 -36.7556 -59.2317 -9.999
1635 -32.3209 -48.1514 -9.999
1640 -22.7274 -41.544 -9.999
1645 -17.6939 -34.8068 -9.999
1660 -9.999 -15.0115 -9.999
1675 -9.999 -9.999 -9.999

27
Figure B-18: Summary of 1 MHz and 1 MHz In-band BeiDou B1 I Bounding Masks

28
Table B-7: 1 MHz and 1 MHz In-band* Galileo E1 BC Bounding Masks (dBm)
Frequency
(MHz)
HPR
1475 -35.2548
1490 -40.5603
1505 -54.5503
1520 -67.9842
1525 -73.9463
1530 -71.4059
1535 -71.6929
1540 -76.3935
1545 -90.805
1550 -93.463
1555 -93.9953
1575* -111.593
1595* -98.7922
1615 -91.4673
1620 -86.9953
1625 -72.9294
1630 -66.8658
1635 -59.6178
1640 -53.3542
1645 -46.7501
1660 -40.3065
1675 -36.0584

29
Figure B-19: Summary of 1 MHz and 1 MHz In-band Galileo E1 BC Bounding Masks

30
Table B-8: 1 MHz and 1 MHz In-band* SBAS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR TIM
1475 -51.0577 -36.3048 -9.999
1490 -63.3312 -36.2106 -9.999
1505 -43.0991 -54.7003 -9.999
1520 -51.2192 -66.2717 -18.1846
1525 -53.1171 -68.445 -17.8801
1530 -52.0402 -70.6184 -19.9122
1535 -52.3345 -72.0554 -21.82
1540 -52.9021 -76.3935 -27.8443
1545 -57.8562 -92.255 -36.8048
1550 -75.2539 -92.7255 -60.008
1555 -83.7667 -92.7703 -64.5788
1575* -118.108 -116.304 -111.015
1595* -71.7432 -98.7297 -65.1566
1615 -74.2469 -91.5423 -45.1318
1620 -66.3442 -82.5362 -46.3982
1625 -54.2824 -66.4259 -37.1343
1630 -56.1713 -61.6283 -29.2349
1635 -43.9075 -55.9303 -22.8593
1640 -43.7996 -44.6667 -18.0685
1645 -43.1834 -48.7722 -14.134
1660 -42.1121 -40.0315 -10.9981
1675 -34.4705 -35.6063 NaN

31
Figure B-20: Summary of 1 MHz and 1 MHz In-band SBAS L1 C/A Bounding Masks
B.1.2 Interference Test Results for 10-MHz LTE

32
Cellular receivers

33
Cellular receivers

34

35

36

37

38

39
Timing receivers
Timing receivers
Using this approach a bounding ITM was produced for each category and signal type
combination. Figure B-33 illustrates how the L1 C/A bounding ITMs compare across the six
categories of receivers. This plot shows the HPR and SPB categories to be the most susceptible
in terms of received interference power levels with the cellular category generally being the most
tolerant of LTE interference.

40
Table B-9: 10 MHz and 10 MHz Intermod* GPS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
1475 -43.3646 -43.0577 -55.2458 -15.2559 -44.9489 -9.999
1490 -39.5303 -46.2115 -60.9099 -16.4053 -56.0659 -9.999
1505 -46.7082 -49.3654 -65.3551 -24.6328 -67.1829 -9.999
1520 -53.6318 -53.6318 -66.3967 -37.0094 -75.2901 -9.999
1525 -55.4429 -55.4429 -70.4213 -42.5001 -75.2341 -12.9574
1530 -60.5293 -60.5293 -72.9934 -51.3793 -75.1091 -15.3651
1530* -73.6293 -75.91 -78.9673 -73.6293 -76.5937 -36.467
1535 -62.445 -62.445 -74.8804 -58.249 -74.501 -17.7728
1540 -63.4002 -63.4002 -81.6489 -62.9739 -77.1892 -28.7517
1545 -72.8009 -70.8562 -89.9247 -72.8009 -79.7538 -38.5472
1550 -81.3979 -79.7539 -93.538 -81.3979 -82.3183 -46.0942
1550* -82.7979 -85.2039 -88.2261 -82.7979 -86.1433 -45.5942
1620 -48.6403 -70.7486 -88.9203 -56.5041 -79.0305 -52.9139
1625 -46.7292 -67.6878 -78.9544 -49.1718 -83.6891 -44.7992
1630 -36.8823 -60.8322 -63.1922 -41.2976 -79.1041 -26.5096
1635 -33.2516 -47.6279 -60.3723 -36.7448 -74.5191 -17.2657
1640 -34.2495 -42.7996 -60.2717 -32.192 -74.405 -15.8067
1645 -34.2708 -43.7334 -61.9748 -27.6393 -74.3188 -13.9425
1660 -36.3256 -41.6121 -53.9123 -18.8014 -72.0909 -13.7382
1675 -44.4325 -38.4705 -49.4567 -16.5372 -64.8224 -15.6612

41
Figure B-33: Summary of 10 MHz and 10MHz Intermod GPS L1 C/A Bounding Masks

42
Table B-10: 10 MHz and 10 MHz Intermod* GPS L1 P Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR SPB CEL
1475 -9.999 -58.2419 -48.2739 -9.999
1490 -9.999 -65.9099 -59.2909 -9.999
1505 -9.999 -65.5172 -70.3079 -9.999
1520 -9.999 -63.7114 -74.0025 -9.999
1525 -12.1984 -65.1348 -75.2341 -12.9574
1530 -14.3977 -64.8352 -75.2091 -15.3651
1530* -27.9178 -68.2977 NaN NaN
1535 -16.5971 -65.3046 -75.151 -17.7728
1540 -19.7263 -66.7822 -75.2145 -20.1804
1545 -29.999 -65.5121 -79.6059 -29.999
1550 -41.6162 -68.946 -85.6683 -29.999
1550* -37.2516 -77.621 NaN NaN
1620 -52.0646 -62.4395 NaN -52.9139
1625 -48.234 -62.7591 -85.8891 -42.7839
1630 -32.0891 -62.2422 -82.0611 -26.5096
1635 -16.731 -61.4098 -78.233 -17.2657
1640 -13.365 -63.4217 -74.405 -15.8067
1645 -9.999 -62.9873 -74.3688 -13.9425
1660 -9.999 -49.5114 -74.2659 -13.7382
1675 -9.999 -46.3192 -65.8474 -15.6612

43
Figure B-34: Summary of 10 MHz and 10 MHz Intermod GPS L1 P Bounding Masks

44
Table B-11: 10 MHz and 10 MHz Intermod* GPS L1 C Bounding Masks (dBm)
Frequency
(MHz)
GLN CEL
1475 -9.999 -9.999
1490 -9.999 -9.999
1505 -9.999 -9.999
1520 -9.999 -9.999
1525 -12.1984 -12.9574
1530 -14.3977 -15.3651
1530* -27.9178 NaN
1535 -16.5971 -17.7728
1540 -19.7263 -20.1804
1545 -29.999 -29.999
1550 -41.6162 -29.999
1550* -37.2516 NaN
1620 -52.0646 -52.9139
1625 -48.234 -42.7839
1630 -32.0891 -26.5096
1635 -16.731 -17.2657
1640 -13.365 -15.8067
1645 -9.999 -13.9425
1660 -9.999 -13.7382
1675 -9.999 -15.6612

45
Figure B-35: Summary of 10 MHz and 10 MHz Intermod GPS L1 C Bounding Masks

46
Table B-12: 10 MHz and 10 MHz Intermod* GLONASS L1 C Bounding Masks (dBm)
Frequency
(MHz)
GAV GLN HPR TIM CEL
1475 -41.5559 -49.0577 -57.1169 -14.6684 -15.1348
1490 -56.5303 -56.5303 -63.9099 -26.6126 -12.2023
1505 -55.7082 -55.7082 -65.1422 -26.4832 -11.4679
1520 -58.6318 -58.6318 -76.9217 -31.4068 -10.7334
1525 -59.1679 -59.1679 -76.8713 -33.7179 -10.4886
1530 -63.5043 -63.5043 -76.7309 -37.5543 -15.4582
1530* -63.4418 -63.4418 -79.6515 -63.4418 -32.967
1535 -64.22 -64.22 -76.5364 -40.595 -20.9174
1540 -66.3502 -66.3502 -76.381 -54.7114 -26.3767
1545 -68.7673 -68.7673 -80.4122 -62.3673 -37.5972
1550 -75.3943 -75.3943 -88.6362 -75.3943 -45.0942
1550* -72.6568 -72.6568 -89.0612 -72.6568 -42.1424
1620 -45.8192 -78.6146 -89.2958 -66.2982 -58.9889
1625 -40.1098 -80.4061 -92.9202 -57.8343 -42.9589
1630 -39.3828 -71.0266 -72.5692 -46.2599 -25.8346
1635 -30.9245 -49.925 -66.0598 -34.1262 -23.6998
1640 -27.7297 -43.7996 -65.2717 -33.0535 -21.5649
1645 -23.4929 -42.2334 -66.1748 -32.0656 -19.43
1660 -16.7459 -39.5371 -59.3114 -25.8014 -17.5882
1675 -9.999 -39.108 -52.4567 -21.0372 -15.177

47
Figure B-36: Summary of 10 MHz and 10 MHz Intermod GLONASS L1 C Bounding
Masks

48
Table B-13: 10 MHz and 10 MHz Intermod* GLONASS L1 P Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR CEL
1475 -9.999 -9.999 -15.1348
1490 -9.999 -11.2502 -12.2023
1505 -12.1713 -16.534 -11.4679
1520 -14.3436 -24.3966 -10.7334
1525 -13.6543 -29.9846 -10.4886
1530 -12.965 -35.0448 -10.2438
1530* -29.5678 -47.5536 -9.999
1535 -12.6165 -38.3146 -9.999
1540 -16.0238 -40.9242 -19.999
1545 -29.999 -43.472 -29.999
1550 -38.8287 -56.6851 -29.999
1550* -38.9037 -56.8351 -29.999
1620 -78.6146 -84.0237 -58.9889
1625 -80.4061 -63.1057 -42.9589
1630 -71.0266 -40.6515 -25.8346
1635 -49.925 -33.9881 -23.6998
1640 -16.666 -28.7876 -21.5649
1645 -19.0538 -24.6255 -19.43
1660 -9.999 -12.3051 -17.5882
1675 -9.999 -9.999 -15.177

49
Figure B-37: Summary of 10 MHz and 10 MHz Intermod GLONASS L1 P Bounding
Masks

50
Table B-14: 10 MHz and 10 MHz Intermod* BeiDou B1 I Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR CEL
1475 -15.2553 -19.5883 -9.999
1490 -18.5174 -27.4185 -9.999
1505 -22.5464 -51.029 -9.999
1520 -28.4227 -65.3777 -9.999
1525 -29.1841 -70.5146 -9.999
1530 -32.0749 -76.2673 -10.4945
1530* NaN -79.8286 -58.9725
1535 -33.6097 -76.6679 -10.9901
1540 -48.4839 -78.8212 -26.0027
1545 -64.2635 -91.9158 -47.9971
1550 -87.2057 -94.6261 -63.9259
1550* NaN -89.1101 -68.4259
1620 -68.727 -77.163 -29.999
1625 -53.1608 -70.407 -29.999
1630 -39.7056 -62.1942 -9.999
1635 -31.4209 -51.7889 -9.999
1640 -27.2274 -42.9315 -9.999
1645 -25.0439 -36.2818 -9.999
1660 -9.999 -14.1615 -9.999
1675 -9.999 -9.999 -9.999

51
Figure B-38: Summary of 10 MHz and 10 MHz Intermod BeiDou B1 I Bounding Masks

52
Table B-15: 10 MHz and 10 MHz Intermod* Galileo E1 BC Bounding Masks (dBm)
Frequency
(MHz)
HPR
1475 -36.1449
1490 -41.2853
1505 -56.8628
1520 -67.5092
1525 -71.2713
1530 -73.8934
1530* -78.9423
1535 -76.5364
1540 -83.6832
1545 -90.83
1550 -91.788
1550* -88.2011
1620 -90.8953
1625 -77.6544
1630 -60.4783
1635 -58.0303
1640 -54.8042
1645 -45.6251
1660 -41.2507
1675 -35.8563

53
Figure B-39: Summary of 10 MHz and 10 MHz Intermod Galileo E1 BC Bounding Masks

54
Table B-16: 10 MHz and 10 MHz Intermod* SBAS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
GLN HPR TIM
1475 -43.0577 -36.9968 NaN
1490 -46.2615 -39.5353 NaN
1505 -49.4654 -56.4753 NaN
1520 -52.6692 -66.8467 -15.9846
1525 -53.0921 -70.5588 -17.8801
1530 -54.2027 -74.2059 -19.9122
1530* -72.4225 -78.9673 -55.4372
1535 -53.0595 -74.9054 -23.82
1540 -58.906 -82.4582 -30.8443
1545 -72.3562 -90.03 -52.5923
1550 -84.7539 -93.013 -63.7205
1550* -81.7289 -88.2261 -64.7455
1620 -68.8442 -90.2953 -53.0107
1625 -61.8949 -77.5544 -49.1718
1630 -53.4838 -60.2533 -39.5099
1635 -47.6279 -56.6803 -26.3593
1640 -42.7996 -49.4667 -21.231
1645 -44.2334 -46.0279 -20.184
1660 -41.6121 -42.1585 -11.4981
1675 -38.4705 -38.5679 NaN

55
Figure B-40: Summary of 10 MHz and 10 MHz Intermod SBAS L1 C/A Bounding Masks
B.1.3 Combined 1 MHz AWGN Test Results and 10 MHz LTE Test Results

56
Figure B-41: Summary of 1&10 MHz, 1 MHz In-band, and 10 MHz Intermod
GPS L1 C/A – HPR Bounding Masks

57
with Certified Aviation GPS L1 C/A Bounding Masks
Note: Certified Aviation Mask has a value of -110 dBm for 1 MHz in band interference

58
GPS L1 P – HPR Bounding Masks

59
GPS L1 P Bounding Masks

60
GPS L1 C - GLN Bounding Masks

61
GPS L1 C Bounding Masks

62
GLONASS L1 C - HPR Bounding Masks

63
GLONASS L1 C Bounding Masks

64
GLONASS L1 P - HPR Bounding Masks

65
GLONASS L1 P Bounding Masks

66
BeiDou B1 I - HPR Bounding Masks

67
BeiDou B1 I Bounding Masks

68
Galileo E1 BC Bounding Masks

69
SBAS L1 C/A - HPR Bounding Masks

70
SBAS L1 C/A Bounding Masks

71
Figure B-56: Summary of Cellular Bounding Masks for
1 & 10 MHz, 1 MHz In-band, and 10 MHz Intermod

72
Figure B-57: Summary of General Aviation Bounding Masks for

73
Figure B-58: Summary of General Location Navigation Bounding Masks for
Figure B-59 compares the ITMs across all emulated GNSS signal types for the HPR receiver
category. The receive interference level from an LTE type signal should not exceed any of the
masks in if all GNSS operation is to be ensured for the HPR category.

74
Figure B-59: Summary of High Precision Bounding Masks for

75
Figure B-60: Summary of Space Based Bounding Masks for

76
Figure B-61: Summary of Timing Bounding Masks for

1
APPENDIX C
LOSS OF LOCK RESULTS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 4
LIST OF TABLES.......................................................................................................................... 5
REFERENCES ............................................................................................................................... 6
APPENDIX..................................................................................................................................... 7
Loss of Lock Results............................................................................................... 7
C.1 Summary of Bounding Masks........................................................................................ 7
C.1.1 GPS L1 C/A............................................................................................................. 7
C.1.2 GPS L1 P ................................................................................................................. 8
C.1.3 GPS L1 C................................................................................................................. 9
C.1.4 GLONASS L1 C.................................................................................................... 10
C.1.5 GLONASS L1 P.................................................................................................... 11
C.1.6 BeiDou B1I............................................................................................................ 12
C.1.7 Galileo E1 BC........................................................................................................ 13
C.1.8 SBAS L1 C/A........................................................................................................ 14
C.2 Summary of Loss-of-Lock Analysis ............................................................................ 15
C.2.1 GPS L1 C/A........................................................................................................... 15
C.2.1.1 Nominal SVs................................................................................................ 15
C.2.1.2 Low Elevation SVs ...................................................................................... 17
C.2.2 GPS L1 P ............................................................................................................... 19
C.2.2.1 Nominal SVs................................................................................................ 19
C.2.3 GPS L1 C............................................................................................................... 24
C.2.3.1 Nominal SVs................................................................................................ 24
C.2.4 GLONASS L1 C.................................................................................................... 29
C.2.4.1 Nominal SVs................................................................................................ 29

3
C.2.5 GLONASS L1 P.................................................................................................... 34
C.2.5.1 Nominal SVs................................................................................................ 34
C.2.6 BeiDou B1I............................................................................................................ 38
C.2.6.1 Nominal SVs................................................................................................ 38
C.2.7 Galileo E1 BC........................................................................................................ 43
C.2.7.1 Nominal SVs................................................................................................ 43
C.2.8 SBAS L1 C/A........................................................................................................ 47
C.2.8.1 Nominal SVs................................................................................................ 47

4
LIST OF FIGURES
Figure C-1: Bounding Masks for GPS L1 C/A with 10 MHz LTE Interference............................ 7
Figure C-2: Bounding Masks for GPS L1 P with 10 MHz LTE Interference................................ 8
Figure C-3: Bounding Masks for GPS L1 C with 10 MHz LTE Interference................................ 9
Figure C-4: Bounding Masks for GLONASS L1 C with 10 MHz LTE Interference................... 10
Figure C-5: Bounding Masks for GLONASS L1 P with 10 MHz LTE Interference................... 11
Figure C-6: Bounding Masks for BeiDou B1I with 10 MHz LTE Interference........................... 12
Figure C-7: Bounding Masks for Galileo E1 BC with 10 MHz LTE Interference....................... 13
Figure C-8: Bounding Masks for SBAS L1 C/A with 10 MHz LTE Interference....................... 14
Figure C-9: Nominal SVs (a) GPS L1 C/A Bounding Masks and (b) GPS L1 C/A Median ...... 16
Figure C-10: Low Elevation SVs (a) GPS L1 C/A Bounding Masks and (b) GPS L1 C/A
Median .......................................................................................................................................... 18
Figure C-11: Nominal SVs (a) GPS L1 P Bounding Masks and (b) GPS L1 P Median............. 21
Figure C-12: Low Elevation SVs (a) GPS L1 P Bounding Masks and (b) GPS L1 P Median ... 23
Figure C-13: Nominal SVs (a) GPS L1 C Bounding Masks and (b) GPS L1 C Median............ 26
Figure C-14: Low Elevation SVs (a) GPS L1 C Bounding Masks and (b) GPS L1 C Median .. 28
Figure C-15: Nominal SVs (a) GLONASS L1 C Bounding Masks and (b) GLONASS L1 C
Median .......................................................................................................................................... 31
Figure C-16: Low Elevation SVs (a) GLONASS L1 C Bounding Masks and (b) GLONASS L1
C Median....................................................................................................................................... 33
Figure C-17: Nominal SVs (a) GLONASS L1 P Bounding Masks and (b) GLONASS L1 P
Median .......................................................................................................................................... 35
Figure C-18: Low Elevation SVs (a) GLONASS L1 P Bounding Masks and (b) GLONASS L1
P Median....................................................................................................................................... 37
Figure C-19: Nominal SVs (a) BeiDou B1I Bounding Masks and (b) BeiDou B1I Median...... 40
Figure C-20: Low Elevation SVs (a) BeiDou B1I Bounding Masks and (b) BeiDou B1I Median
....................................................................................................................................................... 42
Figure C-21: Nominal SVs (a) Galileo E1 BC Bounding Masks and (b) Galileo E1 BC Median
....................................................................................................................................................... 44
Figure C-22: Low Elevation SVs (a) Galileo E1 BC Bounding Masks and (b) Galileo E1 BC
Median .......................................................................................................................................... 46
Figure C-23: Nominal SVs (a) SBAS L1 C/A Bounding Masks and (b) SBAS L1 C/A Median49

5
LIST OF TABLES
Table C-1: Nominal SVs GPS L1 C/A Bounding Masks (dBm).................................................. 15
Table C-2: Nominal SVs GPS L1 C/A Median (dBm)................................................................. 16
Table C-3: Low Elevation SVs GPS L1 C/A Bounding Masks (dBm)........................................ 17
Table C-4: Low Elevation SVs GPS L1 C/A Median (dBm)....................................................... 18
Table C-5: Nominal SVs GPS L1 P Bounding Masks (dBm)...................................................... 19
Table C-6: Nominal SVs GPS L1 P Median (dBm)..................................................................... 20
Table C-7: Low Elevation SVs GPS L1 P Bounding Masks (dBm) ............................................ 22
Table C-8: Low Elevation SVs GPS L1 P Median (dBm) ........................................................... 23
Table C-9: Nominal SVs GPS L1 C Bounding Masks (dBm)...................................................... 24
Table C-10: Nominal SVs GPS L1 C Median (dBm)................................................................... 25
Table C-11: Low Elevation SVs GPS L1 C Bounding Masks (dBm).......................................... 27
Table C-12: Low Elevation SVs GPS L1 P Median (dBm) ......................................................... 28
Table C-13: Nominal SVs GLONASS L1 C Bounding Masks (dBm) ........................................ 29
Table C-14: Nominal SVs GLONASS L1 C Median (dBm) ....................................................... 30
Table C-15: Low Elevation SVs GLONASS L1 C Bounding Masks (dBm)............................... 32
Table C-16: Low Elevation SVs GLONASS L1 C Median (dBm).............................................. 33
Table C-17: Nominal SVs GLONASS L1 P Bounding Masks (dBm)......................................... 34
Table C-18: Nominal SVs GLONASS L1 P Median (dBm)........................................................ 35
Table C-19: Low Elevation SVs GLONASS L1 P Bounding Masks (dBm) ............................... 36
Table C-20: Low Elevation SVs GLONASS L1 P Median (dBm) .............................................. 37
Table C-21: Nominal SVs BeiDou B1I Bounding Masks (dBm)................................................. 38
Table C-22: Nominal SVs BeiDou B1I Median (dBm)................................................................ 39
Table C-23: Low Elevation SVs BeiDou B1I Bounding Masks (dBm)....................................... 41
Table C-24: Low Elevation SVs BeiDou B1I Median (dBm)...................................................... 42
Table C-25: Nominal SVs Galileo E1 BC Bounding Masks (dBm) ............................................ 43
Table C-26: Nominal SVs Galileo E1 BC Median (dBm) ........................................................... 44
Table C-27: Low Elevation SVs Galileo E1 BC Bounding Masks (dBm)................................... 45
Table C-28: Low Elevation SVs Galileo E1 BC Median (dBm).................................................. 46
Table C-29: Nominal SVs SBAS L1 C/A Bounding Masks (dBm)............................................. 47
Table C-30: Nominal SVs SBAS L1 C/A Median (dBm)............................................................ 48

6
REFERENCES

7
APPENDIX
Loss of Lock Results
C.1 Summary of Bounding Masks
C.1.1 GPS L1 C/A
Figure C-1: Bounding Masks for GPS L1 C/A with 10 MHz LTE Interference

8
C.1.2 GPS L1 P
Figure C-2: Bounding Masks for GPS L1 P with 10 MHz LTE Interference

9
C.1.3 GPS L1 C
Figure C-3: Bounding Masks for GPS L1 C with 10 MHz LTE Interference

10
C.1.4 GLONASS L1 C
Figure C-4: Bounding Masks for GLONASS L1 C with 10 MHz LTE Interference

11
C.1.5 GLONASS L1 P
Figure C-5: Bounding Masks for GLONASS L1 P with 10 MHz LTE Interference

12
C.1.6 BeiDou B1I
Figure C-6: Bounding Masks for BeiDou B1I with 10 MHz LTE Interference

13
C.1.7 Galileo E1 BC
Figure C-7: Bounding Masks for Galileo E1 BC with 10 MHz LTE Interference

14
C.1.8 SBAS L1 C/A
Figure C-8: Bounding Masks for SBAS L1 C/A with 10 MHz LTE Interference

15
C.2 Summary of Loss-of-Lock Analysis
C.2.1 GPS L1 C/A
C.2.1.1 Nominal SVs
Loss-of-Lock ITM computed using only “nominally” powered GPS signals
Table C-1: Nominal SVs GPS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
1475 -9.999 -9.999 -31.2458 -9.999 -11.5239 -9.999
1490 -9.999 -9.999 -26.4447 -9.999 -15.602 -9.999
1505 -9.999 -9.999 -39.4615 -9.999 -23.7284 -9.999
1520 -10.7094 -10.2192 -43.3004 -10.7094 -34.2901 -9.999
1525 -19.6251 -10.5456 -44.8713 -19.6251 -44.2341 -9.999
1530 -30.5543 -9.999 -46.7309 -30.5543 -38.2091 -9.999
1535 -35.399 -9.999 -50.8679 -35.399 -45.151 -9.999
1540 -41.3614 -13.7121 -59.1205 -41.3614 -47.3892 -9.999
1545 -50.3634 -19.4403 -66.0247 -50.3634 -59.2506 -9.999
1550 -59.2229 -20.705 -68.8667 -59.2229 -61.4683 -9.999
1620 -9.999 -27.7486 -63.9203 -9.999 -66.5055 -9.999
1625 -9.999 -29.7378 -53.9544 -9.999 -71.4891 -9.999
1630 -9.999 -12.4154 -44.3042 -9.999 -56.4729 -9.999
1635 -9.999 -9.999 -40.1698 -9.999 -64.5191 -9.999
1640 -9.999 -9.999 -37.9451 -9.999 -45.405 -9.999
1645 -9.999 -9.999 -38.1748 -9.999 -42.3688 -9.999
1660 -9.999 -9.999 -37.9123 -9.999 -40.2659 -9.999
1675 -9.999 -9.999 -24.847 -9.999 -17.1974 -9.999

16
Table C-2: Nominal SVs GPS L1 C/A Median (dBm)
Figure C-9: Nominal SVs (a) GPS L1 C/A Bounding Masks and
(b) GPS L1 C/A Median
Frequency
(MHz)
1475 -9.999 -9.999 -9.999 -9.999 -10.7615 -9.999
1490 -9.999 -9.999 -9.999 -9.999 -12.8005 -9.999
1505 -9.999 -9.999 -9.999 -9.999 -16.8637 -9.999
1520 -9.999 -9.999 -13.4422 -9.999 -22.8703 -9.999
1525 -9.999 -9.999 -16.8264 -9.999 -31.8007 -9.999
1530 -9.999 -9.999 -21.2618 -9.999 -32.8139 -9.999
1535 -9.999 -9.999 -26.9407 -9.999 -45.151 -9.999
1540 -9.999 -9.999 -31.4524 -9.999 -45.8019 -9.999
1545 -9.999 -9.999 -34.4658 -9.999 -59.2506 -9.999
1550 -9.999 -9.999 -40.9653 -9.999 -59.3381 -9.999
1620 -9.999 -9.999 -36.5694 -9.999 -56.5085 -9.999
1625 -9.999 -9.999 -31.9266 -9.999 -59.0085 -9.999
1630 -9.999 -9.999 -22.2095 -9.999 -50.5324 -9.999
1635 -9.999 -9.999 -14.6811 -9.999 -64.5191 -9.999
1640 -9.999 -9.999 -10.551 -9.999 -38.9629 -9.999
1645 -9.999 -9.999 -9.999 -9.999 -42.3688 -9.999
1660 -9.999 -9.999 -9.999 -9.999 -26.3502 -9.999
1675 -9.999 -9.999 -9.999 -9.999 -13.5982 -9.999

17
C.2.1.2 Low Elevation SVs
Loss-of-Lock ITM computed using only PRN-24 fixed at -10 dB relative to the nominal received
power levels.
Table C-3: Low Elevation SVs GPS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
1475 -9.999 -19.0577 -48.2458 -9.999 NaN -9.999
1490 -9.999 -25.3312 -30.0449 -9.999 NaN -9.999
1505 -16.6328 -18.724 -41.279 -16.6328 NaN -9.999
1520 -23.7094 -17.6091 -51.9467 -23.7094 NaN -9.999
1525 -32.6251 -22.0921 -54.8713 -32.6251 NaN -9.999
1530 -40.5543 -18.5543 -57.7309 -40.5543 NaN -9.999
1535 -46.399 -23.445 -58.8679 -46.399 NaN -9.999
1540 -52.3614 -24.4252 -69.1205 -52.3614 NaN -9.999
1545 -62.3634 -32.507 -75.0247 -62.3634 NaN -9.999
1550 -70.2229 -41.8765 -78.8667 -70.2229 NaN -9.999
1620 -32.6403 -31.3943 -75.9203 -9.999 NaN -32.1514
1625 -33.7292 -30.7686 -63.9544 -9.999 NaN -9.999
1630 -26.8823 -16.82 -49.9352 -9.999 NaN -12.0846
1635 -9.999 -20.9075 -42.0443 -9.999 NaN -9.999
1640 -9.999 -18.7996 -38.8251 -9.999 NaN -9.999
1645 -9.999 -18.7334 -40.0116 -9.999 NaN -9.999
1660 -9.999 -9.999 -45.9123 -9.999 NaN -9.999
1675 -9.999 -16.4705 -27.6572 -9.999 NaN -9.999

18
Table C-4: Low Elevation SVs GPS L1 C/A Median (dBm)
Figure C-10: Low Elevation SVs (a) GPS L1 C/A Bounding Masks and
(b) GPS L1 C/A Median
Frequency
(MHz)
1475 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1490 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1505 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1520 -9.999 -9.999 -11.8129 -9.999 NaN -9.999
1525 -9.999 -9.999 -17.1529 -9.999 NaN -9.999
1530 -9.999 -9.999 -23.1472 -9.999 NaN -9.999
1535 -9.999 -9.999 -31.1458 -9.999 NaN -9.999
1540 -9.999 -9.999 -32.9976 -9.999 NaN -9.999
1545 -9.999 -9.999 -38.5907 -9.999 NaN -9.999
1550 -9.999 -9.999 -45.0672 -9.999 NaN -9.999
1620 -9.999 -9.999 -38.8637 -9.999 NaN -9.999
1625 -9.999 -9.999 -37.3521 -9.999 NaN -9.999
1630 -9.999 -9.999 -28.4308 -9.999 NaN -9.999
1635 -9.999 -9.999 -16.5783 -9.999 NaN -9.999
1640 -9.999 -9.999 -14.5126 -9.999 NaN -9.999
1645 -9.999 -9.999 -10.2873 -9.999 NaN -9.999
1660 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1675 -9.999 -9.999 -9.999 -9.999 NaN -9.999

19
C.2.2 GPS L1 P
C.2.2.1 Nominal SVs
Table C-5: Nominal SVs GPS L1 P Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 -9.999 -9.999 -9.999
1490 NaN -9.999 -43.5111 -9.999 -9.999 -9.999
1505 NaN -9.999 -50.4615 -9.999 -23.7284 -9.999
1520 NaN -9.999 -53.3004 -9.999 -55.2901 -9.999
1525 NaN -9.999 -52.8098 -17.4427 -61.2341 -9.999
1530 NaN -9.999 -51.2816 NaN NaN -9.999
1535 NaN -9.999 -54.8046 NaN NaN -9.999
1540 NaN -9.999 -54.7822 -37.657 NaN -9.999
1545 NaN -9.999 -56.7246 NaN -59.2506 -9.999
1550 NaN -9.999 -57.6585 -56.4707 -61.4683 -9.999
1620 NaN -9.999 -46.3395 -9.999 -66.5055 -9.999
1625 NaN -9.999 -46.3261 -9.999 -70.4891 -9.999
1630 NaN -12.4154 -45.3042 -9.999 -56.4729 -9.999
1635 NaN -9.999 -44.3723 -9.999 -64.5191 -9.999
1640 NaN -9.999 -42.2717 -9.999 NaN -9.999
1645 NaN -9.999 -36.8869 -9.999 -42.3688 -9.999
1660 NaN -9.999 -29.9614 -9.999 -40.2659 -9.999
1675 NaN -9.999 -26.4567 -9.999 -10.1974 -9.999

20
Table C-6: Nominal SVs GPS L1 P Median (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 -9.999 -9.999 -9.999
1490 NaN -9.999 -42.9099 -9.999 -9.999 -9.999
1505 NaN -9.999 -47.9422 -9.999 -16.8637 -9.999
1520 NaN -9.999 -52.8489 -9.999 -33.3703 -9.999
1525 NaN -9.999 -51.2298 -17.4427 -40.3007 -9.999
1530 NaN -9.999 -49.8352 NaN NaN -9.999
1535 NaN -9.999 -53.2823 NaN NaN -9.999
1540 NaN -9.999 -53.3135 -37.657 NaN -9.999
1545 NaN -9.999 -55.2629 NaN -59.2506 -9.999
1550 NaN -9.999 -57.2322 -56.4707 -59.3381 -9.999
1620 NaN -9.999 -45.2663 -9.999 -66.5055 -9.999
1625 NaN -9.999 -45.2341 -9.999 -58.5085 -9.999
1630 NaN -9.999 -44.1922 -9.999 -56.4729 -9.999
1635 NaN -9.999 -41.1038 -9.999 -64.5191 -9.999
1640 NaN -9.999 -39.0043 -9.999 NaN -9.999
1645 NaN -9.999 -31.9579 -9.999 -42.3688 -9.999
1660 NaN -9.999 -29.6232 -9.999 -40.2659 -9.999
1675 NaN -9.999 -25.847 -9.999 -10.0982 -9.999

21
Figure C-11: Nominal SVs (a) GPS L1 P Bounding Masks and
(b) GPS L1 P Median
power levels.

22
Table C-7: Low Elevation SVs GPS L1 P Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -25.96 NaN NaN -9.999
1490 NaN -9.999 -44.0449 NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.999 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -31.3943 -55.2663 NaN NaN -32.1514
1625 NaN -30.7686 -59.2341 NaN NaN -9.999
1630 NaN -15.4154 NaN -9.999 NaN -12.0846
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 -39.029 NaN NaN -9.999
1660 NaN -9.999 NaN -9.999 NaN -9.999
1675 NaN -9.999 NaN NaN -26.1974 -9.999

23
Table C-8: Low Elevation SVs GPS L1 P Median (dBm)
Figure C-12: Low Elevation SVs (a) GPS L1 P Bounding Masks and
(b) GPS L1 P Median
Frequency
(MHz)
1475 NaN -9.999 -25.96 NaN NaN -9.999
1490 NaN -9.999 -44.0449 NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.77771 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -19.675 -55.2663 NaN NaN -26.0967
1625 NaN -14.0983 -59.2341 NaN NaN -9.999
1630 NaN -13.8262 NaN -9.999 NaN -11.0418
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 -39.029 NaN NaN -9.999
1660 NaN -9.999 NaN -9.999 NaN -9.999
1675 NaN -9.999 NaN NaN -18.0982 -9.999

24
C.2.3 GPS L1 C
C.2.3.1 Nominal SVs
Table C-9: Nominal SVs GPS L1 C Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 NaN NaN NaN -9.999
1490 NaN -9.999 NaN NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.999 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -9.999 NaN NaN NaN -9.999
1625 NaN -9.999 NaN NaN NaN -9.999
1630 NaN -12.4154 NaN NaN NaN -9.999
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 NaN NaN NaN -9.999
1660 NaN -9.999 NaN NaN NaN -9.999
1675 NaN -9.999 NaN NaN NaN -9.999

25
Table C-10: Nominal SVs GPS L1 C Median (dBm)
Frequency
(MHz)
1475 NaN -9.999 NaN NaN NaN -9.999
1490 NaN -9.999 NaN NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.999 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -9.999 NaN NaN NaN -9.999
1625 NaN -9.999 NaN NaN NaN -9.999
1630 NaN -9.999 NaN NaN NaN -9.999
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 NaN NaN NaN -9.999
1660 NaN -9.999 NaN NaN NaN -9.999
1675 NaN -9.999 NaN NaN NaN -9.999

26
Figure C-13: Nominal SVs (a) GPS L1 C Bounding Masks and
(b) GPS L1 C Median
power levels.

27
Table C-11: Low Elevation SVs GPS L1 C Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 NaN NaN NaN -9.999
1490 NaN -9.999 NaN NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.999 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -31.3943 NaN NaN NaN -32.1514
1625 NaN -30.7686 NaN NaN NaN -9.999
1630 NaN -15.4154 NaN NaN NaN -12.0846
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 NaN NaN NaN -9.999
1660 NaN -9.999 NaN NaN NaN -9.999
1675 NaN -9.999 NaN NaN NaN -9.999

28
Table C-12: Low Elevation SVs GPS L1 P Median (dBm)
Figure C-14: Low Elevation SVs (a) GPS L1 C Bounding Masks and
(b) GPS L1 C Median
Frequency
(MHz)
1475 NaN -9.999 NaN NaN NaN -9.999
1490 NaN -9.999 NaN NaN NaN -9.999
1505 NaN -9.999 NaN NaN NaN -9.999
1520 NaN -9.999 NaN NaN NaN -9.999
1525 NaN -9.999 NaN NaN NaN -9.999
1530 NaN -9.999 NaN NaN NaN -9.999
1535 NaN -9.999 NaN NaN NaN -9.999
1540 NaN -9.77771 NaN NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 NaN NaN NaN -9.999
1620 NaN -19.675 NaN NaN NaN -26.0967
1625 NaN -14.0983 NaN NaN NaN -9.999
1630 NaN -13.8262 NaN NaN NaN -11.0418
1635 NaN -9.999 NaN NaN NaN -9.999
1640 NaN -9.999 NaN NaN NaN -9.999
1645 NaN -9.999 NaN NaN NaN -9.999
1660 NaN -9.999 NaN NaN NaN -9.999
1675 NaN -9.999 NaN NaN NaN -9.999

29
C.2.4 GLONASS L1 C
C.2.4.1 Nominal SVs
Table C-13: Nominal SVs GLONASS L1 C Bounding Masks (dBm)
Frequency
(MHz)
1475 -9.999 -17.0577 -22.0199 -9.999 NaN -9.999
1490 -9.999 -9.999 -25.2549 -9.999 NaN -9.999
1505 -9.999 -9.999 -37.9422 -9.999 NaN -9.999
1520 -9.999 -26.2192 -48.9467 -9.999 NaN -9.999
1525 -9.999 -24.0921 -50.8713 -9.999 NaN -9.999
1530 -9.999 -33.0652 -50.7309 -9.999 NaN -9.999
1535 -9.999 -26.897 -52.5364 -9.999 NaN -9.999
1540 -9.999 -28.9021 -53.3135 -9.999 NaN -9.999
1545 -9.999 -29.8816 -54.7246 -9.999 NaN -9.999
1550 -18.6341 -32.884 -56.6585 -9.999 NaN -9.999
1620 -9.999 -42.7896 -66.188 -38.2107 NaN -9.999
1625 -9.999 -34.7686 -56.207 NaN NaN -9.999
1630 -11.4378 -27.82 -57.0348 -17.2599 NaN -9.999
1635 -9.999 -21.9075 -45.1698 -9.999 NaN -9.999
1640 -9.999 -18.7996 -38.3831 -9.999 NaN -9.999
1645 -9.999 -17.7334 -37.8222 -9.999 NaN -9.999
1660 -9.999 -18.6121 -31.6007 -9.999 NaN -9.999
1675 -9.999 -9.999 -23.2063 -9.999 NaN -9.999

30
Table C-14: Nominal SVs GLONASS L1 C Median (dBm)
Frequency
(MHz)
1475 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1490 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1505 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1520 -9.999 -9.999 -16.3549 -9.999 NaN -9.999
1525 -9.999 -9.999 -16.3513 -9.999 NaN -9.999
1530 -9.999 -9.999 -20.7887 -9.999 NaN -9.999
1535 -9.999 -9.999 -25.2012 -9.999 NaN -9.999
1540 -9.999 -9.999 -23.0492 -9.999 NaN -9.999
1545 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1550 -9.999 -9.999 -38.8832 -9.999 NaN -9.999
1620 -9.999 -15.0761 -53.4166 -9.999 NaN -9.999
1625 -9.999 -9.999 -37.9896 NaN NaN -9.999
1630 -9.999 -9.999 -33.3659 -9.999 NaN -9.999
1635 -9.999 -9.999 -23.4376 -9.999 NaN -9.999
1640 -9.999 -9.999 -15.6777 -9.999 NaN -9.999
1645 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1660 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1675 -9.999 -9.999 -9.999 -9.999 NaN -9.999

31
Figure C-15: Nominal SVs (a) GLONASS L1 C Bounding Masks and
(b) GLONASS L1 C Median
Loss-of-Lock ITM computed using only SV-01 fixed at -10 dB relative to the nominal received
power levels.

32
Table C-15: Low Elevation SVs GLONASS L1 C Bounding Masks (dBm)
Frequency
(MHz)
1475 -9.999 -27.0577 -25.4968 -9.999 NaN -9.999
1490 -13.7647 -13.7647 -32.0449 -9.999 NaN -9.999
1505 -9.999 -9.999 -51.0301 -9.999 NaN -9.999
1520 -11.8154 -36.2192 -61.9467 -9.999 NaN -9.999
1525 -10.771 -35.0921 -62.8713 -9.999 NaN -9.999
1530 -17.7766 -43.0652 -63.7309 -11.7766 NaN -9.999
1535 -11.722 -42.897 -64.5364 -11.722 NaN -9.999
1540 -9.999 -47.9021 -62.3935 -9.999 NaN -20.9017
1545 -19.6831 -45.8816 -62.2675 -18.6831 NaN -9.999
1550 -27.2693 -50.884 -63.138 -35.7205 NaN -9.999
1620 -18.6707 -52.7896 -77.7669 -49.2107 NaN -46.1514
1625 -9.999 -59.7686 -66.207 -34.1718 NaN -21.579
1630 -15.4378 -33.7641 -59.1922 -24.2599 NaN -18.0846
1635 -9.999 -28.9075 -49.3014 -9.999 NaN -14.2157
1640 -9.999 -26.7996 -40.194 -9.999 NaN -13.0403
1645 -9.999 -25.7334 -39.8279 -9.999 NaN -12.005
1660 -9.999 -27.6121 -33.6007 -9.999 NaN -10.9311
1675 -9.999 -18.4705 -29.4429 -9.999 NaN -9.999

33
Table C-16: Low Elevation SVs GLONASS L1 C Median (dBm)
Figure C-16: Low Elevation SVs (a) GLONASS L1 C Bounding Masks and
(b) GLONASS L1 C Median
Frequency
(MHz)
1475 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1490 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1505 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1520 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1525 -9.999 -9.999 -11.1529 -9.999 NaN -9.999
1530 -9.999 -9.999 -15.1472 -9.999 NaN -9.999
1535 -9.999 -9.999 -23.0548 -9.999 NaN -9.999
1540 -9.999 -9.999 -26.9976 -9.999 NaN -9.999
1545 -18.6831 -9.999 -33.3833 -9.999 NaN -9.999
1550 -19.6341 -19.0674 -44.8832 -14.8166 NaN -9.999
1620 -9.999 -27.3583 -67.4381 -9.999 NaN -27.5419
1625 -9.999 -14.064 -48.9016 -9.999 NaN -15.5228
1630 -9.999 -10.0739 -36.8036 -9.999 NaN -10.2921
1635 -9.999 -9.999 -28.9246 -9.999 NaN -11.1461
1640 -9.999 -9.999 -19.5126 -9.999 NaN -10.7826
1645 -9.999 -9.999 -12.6969 -9.999 NaN -10.2662
1660 -9.999 -9.999 -9.999 -9.999 NaN -9.999
1675 -9.999 -9.999 -9.999 -9.999 NaN -9.999

34
C.2.5 GLONASS L1 P
C.2.5.1 Nominal SVs
Table C-17: Nominal SVs GLONASS L1 P Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -9.999 NaN NaN -9.999
1525 NaN -9.999 -10.3721 NaN NaN -9.999
1530 NaN -9.999 -13.2698 NaN NaN -9.999
1535 NaN -9.999 -17.1146 NaN NaN -9.999
1540 NaN -9.999 -23.0492 NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 -37.6101 NaN NaN -9.999
1620 NaN -42.7896 -59.0987 NaN NaN -9.999
1625 NaN -34.7686 -40.7037 NaN NaN -9.999
1630 NaN -22.4154 -30.8015 NaN NaN -9.999
1635 NaN -11.24 -24.9246 NaN NaN -9.999
1640 NaN -9.999 -16.5126 NaN NaN -9.999
1645 NaN -9.999 -12.5755 NaN NaN -9.999
1660 NaN -9.999 -9.999 NaN NaN -9.999
1675 NaN -9.999 -9.999 NaN NaN -9.999

35
Table C-18: Nominal SVs GLONASS L1 P Median (dBm)
Figure C-17: Nominal SVs (a) GLONASS L1 P Bounding Masks and
(b) GLONASS L1 P Median
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -9.999 NaN NaN -9.999
1525 NaN -9.999 -9.999 NaN NaN -9.999
1530 NaN -9.999 -9.999 NaN NaN -9.999
1535 NaN -9.999 -11.5269 NaN NaN -9.999
1540 NaN -9.999 -13.4983 NaN NaN -9.999
1545 NaN -9.999 NaN NaN NaN -9.999
1550 NaN -9.999 -30.364 NaN NaN -9.999
1620 NaN -21.685 -49.7287 NaN NaN -9.999
1625 NaN -14.0983 -36.9373 NaN NaN -9.999
1630 NaN -11.1903 -25.1865 NaN NaN -9.999
1635 NaN -9.999 -16.2811 NaN NaN -9.999
1640 NaN -9.999 -12.4209 NaN NaN -9.999
1645 NaN -9.999 -9.999 NaN NaN -9.999
1660 NaN -9.999 -9.999 NaN NaN -9.999
1675 NaN -9.999 -9.999 NaN NaN -9.999

36
Loss-of-Lock ITM computed using only SV-01 fixed at -10 dB relative to the nominal received
power levels.
Table C-19: Low Elevation SVs GLONASS L1 P Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -11.4466 NaN NaN -9.999
1525 NaN -9.999 -14.3721 NaN NaN -9.999
1530 NaN -9.999 -19.2698 NaN NaN -9.999
1535 NaN -9.999 -26.1146 NaN NaN -9.999
1540 NaN -9.999 -29.0492 NaN NaN -9.999
1545 NaN -9.999 -32.772 NaN NaN -9.999
1550 NaN -18.5007 -49.6101 NaN NaN -9.999
1620 NaN -52.7896 -71.0987 NaN NaN -46.1514
1625 NaN -59.7686 -53.7037 NaN NaN -21.579
1630 NaN -33.7641 -34.8015 NaN NaN -18.0846
1635 NaN -22.8 -28.9246 NaN NaN -14.2157
1640 NaN -9.999 -20.8429 NaN NaN -13.0403
1645 NaN -9.999 -16.5755 NaN NaN -12.005
1660 NaN -9.999 -9.999 NaN NaN -10.9311
1675 NaN -9.999 -9.999 NaN NaN -9.999

37
Table C-20: Low Elevation SVs GLONASS L1 P Median (dBm)
Figure C-18: Low Elevation SVs (a) GLONASS L1 P Bounding Masks and
(b) GLONASS L1 P Median
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -9.999 NaN NaN -9.999
1525 NaN -9.999 -11.5759 NaN NaN -9.999
1530 NaN -9.999 -9.999 NaN NaN -9.999
1535 NaN -9.999 -16.5269 NaN NaN -9.999
1540 NaN -9.999 -22.4202 NaN NaN -9.999
1545 NaN -9.999 -31.4723 NaN NaN -9.999
1550 NaN -9.999 -36.364 NaN NaN -9.999
1620 NaN -36.371 -57.7287 NaN NaN -45.6181
1625 NaN -24.7626 -45.4373 NaN NaN -21.3128
1630 NaN -18.2821 -31.1865 NaN NaN -14.3349
1635 NaN -11.74 -21.7009 NaN NaN -13.2545
1640 NaN -9.999 -15.2558 NaN NaN -12.3033
1645 NaN -9.999 -10.4087 NaN NaN -11.2692
1660 NaN -9.999 -9.999 NaN NaN -10.4651
1675 NaN -9.999 -9.999 NaN NaN -9.91455

38
C.2.6 BeiDou B1I
C.2.6.1 Nominal SVs
Table C-21: Nominal SVs BeiDou B1I Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -26.279 NaN NaN -9.999
1520 NaN -9.999 -45.1027 NaN NaN -9.999
1525 NaN NaN -50.0021 NaN NaN -9.999
1530 NaN -9.999 -52.9673 NaN NaN -9.999
1535 NaN -9.999 -56.8679 NaN NaN -9.999
1540 NaN -10.1664 -58.8462 NaN NaN -9.999
1545 NaN -9.999 -68.8158 NaN NaN -9.999
1550 NaN -52.2057 -76.7261 NaN NaN -9.999
1620 NaN -30.752 -56.188 NaN NaN -9.999
1625 NaN -9.999 -54.207 NaN NaN -9.999
1630 NaN -9.999 -42.2442 NaN NaN -9.999
1635 NaN -9.999 -33.3014 NaN NaN -9.999
1640 NaN -9.999 -24.194 NaN NaN -9.999
1645 NaN -9.999 -16.1068 NaN NaN -9.999
1660 NaN -9.999 -9.999 NaN NaN -9.999
1675 NaN -9.999 -9.999 NaN NaN -9.999

39
Table C-22: Nominal SVs BeiDou B1I Median (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -9.999 NaN NaN -9.999
1525 NaN NaN -9.999 NaN NaN -9.999
1530 NaN -9.999 -9.999 NaN NaN -9.999
1535 NaN -9.999 -9.999 NaN NaN -9.999
1540 NaN -10.1664 -19.99 NaN NaN -9.999
1545 NaN -9.999 -43.4035 NaN NaN -9.999
1550 NaN -52.2057 -69.2392 NaN NaN -9.999
1620 NaN -30.752 -9.999 NaN NaN -9.999
1625 NaN -9.999 -9.999 NaN NaN -9.999
1630 NaN -9.999 -9.999 NaN NaN -9.999
1635 NaN -9.999 -9.999 NaN NaN -9.999
1640 NaN -9.999 -9.999 NaN NaN -9.999
1645 NaN -9.999 -9.999 NaN NaN -9.999
1660 NaN -9.999 -9.999 NaN NaN -9.999
1675 NaN -9.999 -9.999 NaN NaN -9.999

40
Figure C-19: Nominal SVs (a) BeiDou B1I Bounding Masks and
(b) BeiDou B1I Median
power levels.

41
Table C-23: Low Elevation SVs BeiDou B1I Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -18.2685 NaN NaN -9.999
1505 NaN -9.999 -41.279 NaN NaN -9.999
1520 NaN -9.999 -57.1027 NaN NaN -9.999
1525 NaN -9.999 -60.0021 NaN NaN -9.999
1530 NaN -13.3999 -64.9673 NaN NaN NaN
1535 NaN -19.2972 -66.8679 NaN NaN NaN
1540 NaN -22.3339 -68.8462 NaN NaN NaN
1545 NaN -38.2885 -80.8158 NaN NaN NaN
1550 NaN -70.2057 -80.8683 NaN NaN NaN
1620 NaN -32.752 -66.188 NaN NaN NaN
1625 NaN -18.8299 -65.207 NaN NaN NaN
1630 NaN -11.8523 -52.2442 NaN NaN NaN
1635 NaN -9.999 -42.3014 NaN NaN NaN
1640 NaN -9.999 -32.194 NaN NaN NaN
1645 NaN -9.999 -27.1068 NaN NaN NaN
1660 NaN -9.999 -9.999 NaN NaN NaN
1675 NaN -9.999 -9.999 NaN NaN NaN

42
Table C-24: Low Elevation SVs BeiDou B1I Median (dBm)
Figure C-20: Low Elevation SVs (a) BeiDou B1I Bounding Masks and
(b) BeiDou B1I Median
Frequency
(MHz)
1475 NaN -9.999 -9.999 NaN NaN -9.999
1490 NaN -9.999 -9.999 NaN NaN -9.999
1505 NaN -9.999 -9.999 NaN NaN -9.999
1520 NaN -9.999 -9.999 NaN NaN -9.999
1525 NaN -9.999 -9.999 NaN NaN -9.999
1530 NaN -13.3999 -10.1063 NaN NaN NaN
1535 NaN -19.2972 -13.0842 NaN NaN NaN
1540 NaN -22.3339 -26.49 NaN NaN NaN
1545 NaN -38.2885 -54.9035 NaN NaN NaN
1550 NaN -70.2057 -79.9274 NaN NaN NaN
1620 NaN -32.752 -44.695 NaN NaN NaN
1625 NaN -18.8299 -9.999 NaN NaN NaN
1630 NaN -11.8523 -17.801 NaN NaN NaN
1635 NaN -9.999 -9.999 NaN NaN NaN
1640 NaN -9.999 -9.999 NaN NaN NaN
1645 NaN -9.999 -9.999 NaN NaN NaN
1660 NaN -9.999 -9.999 NaN NaN NaN
1675 NaN -9.999 -9.999 NaN NaN NaN

43
C.2.7 Galileo E1 BC
C.2.7.1 Nominal SVs
Table C-25: Nominal SVs Galileo E1 BC Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN NaN -21.0199 NaN NaN NaN

44
Table C-26: Nominal SVs Galileo E1 BC Median (dBm)
Figure C-21: Nominal SVs (a) Galileo E1 BC Bounding Masks and
(b) Galileo E1 BC Median
Frequency
(MHz)

45
power levels.
Table C-27: Low Elevation SVs Galileo E1 BC Bounding Masks (dBm)
Frequency
(MHz)

46
Table C-28: Low Elevation SVs Galileo E1 BC Median (dBm)
Figure C-22: Low Elevation SVs (a) Galileo E1 BC Bounding Masks and
(b) Galileo E1 BC Median
Frequency
(MHz)

47
C.2.8 SBAS L1 C/A
C.2.8.1 Nominal SVs
Table C-29: Nominal SVs SBAS L1 C/A Bounding Masks (dBm)
Frequency
(MHz)
1475 NaN -17.0577 -23.0199 -9.999 NaN NaN
1490 NaN -23.3312 -25.2549 -9.999 NaN NaN
1505 NaN -21.4491 -32.9753 -9.999 NaN NaN
1520 NaN -17.6091 -41.9467 -9.999 NaN NaN
1525 NaN -23.0921 -44.8713 -9.999 NaN NaN
1530 NaN -27.0652 -46.7309 -9.999 NaN NaN
1535 NaN -29.897 -50.8679 -9.999 NaN NaN
1540 NaN -30.9021 -56.3935 -9.999 NaN NaN
1545 NaN -36.8816 -62.2675 -9.999 NaN NaN
1550 NaN -55.7539 -64.7261 -35.7205 NaN NaN
1620 NaN -40.7896 -65.9203 -9.999 NaN NaN
1625 NaN -31.7686 -51.9544 -9.999 NaN NaN
1630 NaN -25.4838 -39.9352 -9.999 NaN NaN
1635 NaN -19.6279 -38.4688 -9.999 NaN NaN
1640 NaN -11.7996 -37.9451 -9.999 NaN NaN
1645 NaN -14.7334 -35.8279 -9.999 NaN NaN
1660 NaN -10.3055 -29.4585 -9.999 NaN NaN
1675 NaN -9.999 -22.2063 -9.999 NaN NaN

48
Table C-30: Nominal SVs SBAS L1 C/A Median (dBm)
Frequency
(MHz)
1475 NaN -9.999 -9.999 -9.999 NaN NaN
1490 NaN -9.999 -9.999 -9.999 NaN NaN
1505 NaN -9.999 -9.999 -9.999 NaN NaN
1520 NaN -9.999 -11.8986 -9.999 NaN NaN
1525 NaN -9.999 -16.8264 -9.999 NaN NaN
1530 NaN -9.999 -21.6315 -9.999 NaN NaN
1535 NaN -10.1481 -27.116 -9.999 NaN NaN
1540 NaN -14.1664 -31.3096 -9.999 NaN NaN
1545 NaN -9.999 -34.9536 -9.999 NaN NaN
1550 NaN -36.9631 -40.9457 -35.7205 NaN NaN
1620 NaN -9.999 -39.4386 -9.999 NaN NaN
1625 NaN -9.999 -32.9266 -9.999 NaN NaN
1630 NaN -11.6903 -23.4623 -9.999 NaN NaN
1635 NaN -9.999 -14.4376 -9.999 NaN NaN
1640 NaN -9.999 -10.8429 -9.999 NaN NaN
1645 NaN -9.999 -9.999 -9.999 NaN NaN
1660 NaN -9.999 -9.999 -9.999 NaN NaN
1675 NaN -9.999 -9.999 -9.999 NaN NaN

49
Figure C-23: Nominal SVs (a) SBAS L1 C/A Bounding Masks and
(b) SBAS L1 C/A Median

1
APPENDIX D
CONDUCTED TEST DETAILS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 4
REFERENCES ............................................................................................................................... 5
APPENDIX..................................................................................................................................... 6
Conducted Test....................................................................................................... 6
D.1 OOBE Level Settings..................................................................................................... 6

3
LIST OF FIGURES
No table of figures entries found.

4
LIST OF TABLES
Table D-1: Ratio of OOBE limit density to LTE power for setting OOBE testing levels ............. 6
Table D-2: OOBE levels (in dBm/MHz) for the LTE tests ranges................................................. 6

6
APPENDIX
Conducted Test
D.1 OOBE Level Settings
The key points associated with wired testing were largely covered in the main body of the report
or in the previous section with review of SPIGAT. One area that generated much discussion was
simulation of FCC and Proposed OOBE limits in the wired testing. The OOBE levels to be
generated were determined by assuming a fixed ratio between OOBE limit and LTE signal power
for each of the four cases (from the combination of transmitter type and OOBE limit sponsor) as
summarized in Table D-1. The appropriate ratio was applied (added) to the target LTE signal
power at each point in the test to determine the corresponding OOBE level to generate. Each ratio
was determined by assuming OOBE is at the limit when LTE power levels are at the specified
maximums of 62 dBm (32 dBW) for base stations and 23 dBm (-7 dBW) for handsets. All limits
were taken without consideration for sloping specified within (or any levels specified outside) the
RNSS band. Therefore, the FCC limit was taken as -40 dBm/MHz for both handsets and base
stations and the Proposed limits was taken as -70 dBm/MHz for base stations and -75 dBm/MHz
for handsets.
Table D-1: Ratio of OOBE limit density to LTE power for setting OOBE testing levels
OOBE density LTE power ratio OOBE/LTE
[dBm/MHz] [dBm] [dB/MHz]
FCC base station -40 62 -102
FCC handset -40 23 -63
Proposed base station -70 62 -132
Proposed handset -75 23 -98
The appropriate ratio was applied to the target LTE signal power at each point in the test to
determine the corresponding OOBE level to generate. For example, the LTE uplink at 1630 MHz
was tested in a power range from -80 to -10 dBm. For the FCC limit (adding the ratio -63 dB/MHz),
the OOBE range was -143 to -73 dBm/MHz (-233 to -163 dBW/Hz). For the Proposed limit
(adding the ratio -98 dB/MHz), the corresponding OOBE range was -178 to -108 dBm/MHz (-268
to -198 dBW/Hz). Table D-2 shows the OOBE levels generated for each of the LTE test ranges.
Table D-2: OOBE levels (in dBm/MHz) for the LTE tests ranges
FCC Proposed
LTE power OOBE density OOBE density
[dBm] [dBm/MHz] [dBm/MHz]
base
station 1475 -80 To -10 -182 to -112 -212 to -142

7
1490 -80 To -10 -182 to -112 -212 to -142
1505 -80 to -10 -182 to -112 -212 to -142
1520 -80 to -10 -182 to -112 -212 to -142
1525 -80 to -10 -182 to -112 -212 to -142
1530 -80 to -10 -182 to -112 -212 to -142
1535 -80 to -10 -182 to -112 -212 to -142
1540 -80 to -10 -182 to -112 -212 to -142
1545 -100 to -30 -202 to -132 -232 to -162
1550 -100 to -30 -202 to -132 -232 to -162
handset 1620 -100 to -30 -163 to -93 -198 to -128
1625 -100 to -30 -163 to -93 -198 to -128
1630 -80 to -10 -143 to -73 -178 to -108
1635 -80 to -10 -143 to -73 -178 to -108
1640 -80 to -10 -143 to -73 -178 to -108
1645 -80 to -10 -143 to -73 -178 to -108
1660 -80 to -10 -143 to -73 -178 to -108
base
station 1675 -80 to -10 -182 to -112 -212 to -142

1
APPENDIX E
ANTENNA CHARACTERIZATION DETAILS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 5
REFERENCES ............................................................................................................................... 6
APPENDIX..................................................................................................................................... 7
Antenna Characterization Details ........................................................................... 7
E.1 Anechoic Chamber Measurements................................................................................. 7

3
LIST OF FIGURES
Figure E-1: VPOL Gain Patterns for 14 External Antennas (1475 MHz)...................................... 7
Figure E-2: HPOL Gain Patterns for 14 External Antennas (1475 MHz)...................................... 8
Figure E-3: VPOL Gain Patterns for 14 External Antennas (1490 MHz)...................................... 9
Figure E-4: HPOL Gain Patterns for 14 External Antennas (1490 MHz).................................... 10
Figure E-5: VPOL Gain Patterns for 14 External Antennas (1495 MHz).................................... 11
Figure E-10: HPOL Gain Patterns for 14 External Antennas (1520 MHz).................................. 16
Figure E-11: VPOL Gain Patterns for 14 External Antennas (1530 MHz).................................. 17
Figure E-25: RHCP Gain Patterns for 14 External Antennas (1575 MHz).................................. 31

4

5
LIST OF TABLES

7
APPENDIX
Antenna Characterization Details
E.1 Anechoic Chamber Measurements
Figure E-1: VPOL Gain Patterns for 14 External Antennas (1475 MHz)
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

8
Figure E-2: HPOL Gain Patterns for 14 External Antennas (1475 MHz)
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

9
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

10
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

11
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

12
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

13
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

14
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

15
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

16
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

17
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

18
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

19
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

20
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

21
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

22
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

23
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

24
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

25
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

26
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

27
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

28
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

29
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

30
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

31
Figure E-25: RHCP Gain Patterns for 14 External Antennas (1575 MHz)
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

32
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

33
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

34
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

35
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

36
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

37
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

38
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

39
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

40
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

41
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

42
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

43
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

44
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

45
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

46
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

47
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

48
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

49
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

50
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

51
-150 -100 -50 0 50 100 150
-60
-40
-20
0
20
40
60
Gain
(dBi)

1
APPENDIX F
PROPAGATION PATH AND AGGREGATE EFFECT
COMPUTATION MODELS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 4
LIST OF TABLES.......................................................................................................................... 5
REFERENCES ............................................................................................................................... 6
APPENDIX..................................................................................................................................... 8
Propagation Path and Aggregate Effect Computation Models ............................... 8
F.1 Propagation Path Model General Aspects...................................................................... 8
F.2 Single Path Model .......................................................................................................... 8
F.2.1 Single Path Median Isotropic Propagation Model................................................... 8
F.2.1.1 Short Range Path Segment (Two-Ray) Isotropic Model ............................... 9
F.2.1.1.1 Short Range Path Segment (Level-Terrain Two-Ray) Isotropic Model........ 9
F.2.1.1.2 Sloping Terrain Two-Ray Isotropic Path Model ......................................... 11
F.2.1.1.3 Two-Ray Path Model Aspects for Directive, Dual-Polarization Source
Antennas 13
F.2.1.2 Medium Range Path Segment Erceg-Greenstein Isotropic Model.............. 15
F.2.1.3 Medium Range Path Segment Exponential Fit Isotropic Model ................. 16
F.2.1.4 Long Range Segment Path Isotropic Model ................................................ 16
F.2.1.5 Free Space Path Isotropic Model ................................................................. 19
F.2.2 Single Path Probabilistic Propagation Model ........................................................ 19
F.2.2.1 Single Path Probabilistic Propagation Model Basic Aspects....................... 19
F.2.2.2 Median Path Segment Propagation Parameters and Segment Break Points 20
F.2.2.3 Single Path Probabilistic Propagation Slow Fading Parameters, µ and σ... 22
F.2.2.4 Single Path Probabilistic Propagation Fast-Fading Parameters, L, ψ0, ρ0... 23
F.3 Aggregate Propagation Effects Computation Methods................................................ 25
F.3.1 Randomly-Located Source Statistics ..................................................................... 25
F.3.1.1 Randomly-Located Source Mean Aggregate Interference Power ............... 25

3
F.3.1.2 Randomly-Located Source Aggregate Interference Power Standard
Deviation 25
F.3.1.3 Randomly-Located Source Aggregate Interference Power Probability
Distribution..................................................................................................................... 26
F.3.2 Discretely-Located Source Statistics ..................................................................... 27
F.3.2.1 Discrete Source Mean Aggregate Interference Power................................. 27
F.3.2.2 Discrete Source Aggregate Interference Power Probability Distribution.... 28
F.3.2.2.1 General Discrete Source Aggregate Interference Power Probability
Distribution..................................................................................................................... 28
F.3.2.2.2 Discrete Source Aggregate Characteristic Function with Sectorization...... 28

4
LIST OF FIGURES
Figure F-1: Level-Terrain Two-Ray Path Model Geometry......................................................... 10
Figure F-2: Example Level Terrain Two-Ray Isotropic Median Path Spreading Factor ............. 11
Figure F-3: Sloping Terrain Two-Ray Path Model Geometry...................................................... 12
Figure F-4: Example Erceg-Greenstein Isotropic Median Path Spreading Factor ....................... 16
Figure F-5: Example Hata-Okumura Median Isotropic Path Spreading Factor ........................... 17
Figure F-6: Okumura Slope Correction Factor Data Points and Fit Equations............................ 18

6
REFERENCES
[1] M. Hata, "Empirical formula for propagation loss in land mobile radio services," IEEE
Trans. Veh. Technol., vol. 29, pp. 317-325, Aug. 1980.
[2] Y. Okumura, E. Ohmor, T. Kawano and K. Fukua, "Field strength and its variability in
UHF and VHF land-mobile radio service," Review of Electrical Communications Lab., vol.
16, no. 9, 1968.
[3] D. Parsons, The Mobile Radio Propagation Channel, Chichester, England: John Wiley &
Sons, 1996.
[4] V. Erceg, L. J. Greenstein, S. Y. Tjandra, S. R. Parkoff, A. Gupta, B. Kulic, A. Julius and
R. Bianchi, "An Empirically Based Path Loss Model for Wireless Channels in Suburban
Environments," IEEE Journal on Selected Areas in Communications, vol. 17, no. 7, July
1999.
[5] Status Report: Assessment of Compatibility of Planned LightSquared Ancillary Terrestrial
Component Transmissions in the 1526-1536 MHz Band with Certified Aviation GPS
Receivers, FAA Report PR 25, January 25, 2012.
[6] Handout for wireless comm. graduate course, University of Illinois (ECE559), Professor V.
V. Veeraldi pp. 53-56.
[7] C. Loo, "A Statistical Model for a Land Mobile Satellite Link," IEEE Transactions on
Vehicle Technology, vol. 34, no. 3, Aug. 1985.
[8] K. M. Peterson and R. J. Erlandson, "Analytic Statistical Model for Aggregate Radio
Frequency Interference to Airborne GPS Receivers from Ground-Based Emitters," Journal
of the Institute of Navigation, vol. 59, no. 1, Spring, 2012.
to Airborne GPS Receivers from Ground Based Emitters (DOT/FAA/TC-14/30, Rev. 1),
August 7, 2017.
[10] J. A. Gubner, "A New Formula for Lognormal Characteristic Function," IEEE
Transactions on Vehicle Technology, vol. 55, no. 5, pp. 1668- 1671, Sept. 2006.
[11] K. Chung, A Course In Probability Theory, New York: Academic Press, 1974.

8
APPENDIX
Propagation Path and Aggregate Effect Computation Models
F.1 Propagation Path Model General Aspects
All the RFI propagation path models used herein are based on the flat-earth approximation. In
other words, the ground under the aircraft is assumed to be essentially smooth and flat out to a
radio horizon from the point on the ground directly under the aircraft. In line-of-sight
propagation conditions at radio frequencies near the GPS carrier, this radio horizon value
depends, in general, on the aircraft GNSS and RFI source antenna heights and the amount of
atmospheric refraction along the propagation path. A 4/3 Earth radius approximation for the
refractive effect on the radio horizon will be used in all propagation models.
The propagation model described in this document basically consists of two different types of
models. Those scenarios where blockage is a factor are typically evaluated using probabilistic
models while clear line-of-sight scenarios are analyzed using free space path loss.
For scenarios that involve environments in which emissions from adjacent channel systems
undergo scattering, reflections, and absorption prior to arriving at an aircraft's GPS antenna, a
probabilistic propagation model must be used. These propagation environments were the subject
of much research during the development of cellular radio systems. As a result, the cellular radio
community has over the years developed propagation models which for the most part are
applicable to the present RFI scenarios under consideration.
Generally, in this document the point at which free space propagation path model is used occurs
at an aircraft antenna height of 550 meters. Above 550 meters, various parameter limits
associated with the probabilistic models are exceeded thereby making the model use
problematic. Also at these aircraft heights, line-of-sight conditions generally prevail which
means that free space path loss is the most appropriate model to apply.
Below the 550 meter aircraft antenna height the probabilistic models developed by the cellular
radio community become applicable. Each of these models have one feature in common; the
probabilistic nature of the propagation path model is well approximated by the product of a slow
fading process and a fast fading process. The slow fading process is very well approximated by a
log-normal distribution while the fast fading process is described by a non-central Chi-Squared
distribution. The log-normal component is completely determined by two parameters, µ and σ
and the Chi-Squared process by the parameters L, ψo and ρo. The range dependent median path
model between the GPS antenna and the interference source determines the parameter µ while
the remaining parameters vary with range depending on the scenario. In some special cases
below 550 m aircraft antenna height, free-space (1/r2
) propagation is assumed.
F.2 Single Path Model
F.2.1 Single Path Median Isotropic Propagation Model

9
The single path median isotropic propagation model used depends upon the GPS aircraft antenna
height and emitter source antenna height above ground. For aircraft antenna heights greater than
or equal to 550 meters, a free space propagation model is used. The median isotropic propagation
path model for aircraft antenna heights below 550 meters consists in general of three segments,
depending on the scenario, appropriately blended together. They form a continuous function of
the lateral separation radius r (in meters) between the aircraft and RFI source antennas. At the
shorter radial distances, conventional two-ray propagation is used. That model assumes a direct
path between the two antennas interacts with one other path reflected from the ground between
antennas. For radii greater than about 1 km, the Hata-Okumura suburban propagation model (
[1], [2]) is used. At radii beyond 20 km, a long distance extension is made to the standard Hata-
Okumura formulation. If necessary to achieve propagation path model continuity, the two-ray
and Hata-Okumura model segments are connected together by one of two means. One method
uses a single, constant slope exponential fit function that covers radius values in between
segments. The other method, used in mobile broadband handset station scenarios with aircraft
antenna heights below 80 meters and source antennas less than 2 meters, is a modified Erceg-
Greenstein path loss segment. Since median isotropic propagation path model is generally a
function of the aircraft and RFI source antenna heights, the path loss is adapted to each specific
RFI scenario.
For convenience, this Appendix generally uses the term "path spreading factor" (PF), which is
the algebraic reciprocal of the path loss.
F.2.1.1 Short Range Path Segment (Two-Ray) Isotropic Model
F.2.1.1.1 Short Range Path Segment (Level-Terrain Two-Ray) Isotropic Model
Measurements indicate that at short ranges, where few obstructions are present, the GPS antenna
receives interference from an RFI source by both a direct and reflected path [3]. On the reflected
path, the electromagnetic wave is reflected from the earth's surface with a range-dependent
complex reflection coefficient ρν(r)1
. In addition to the relative phase delay between the direct
and reflected rays, the complex reflection coefficient both attenuates and adds phase shift to the
reflected ray. When the two rays additively combine at the receive antenna, the resulting
received power fluctuates with lateral separation range r. Figure F-1 illustrates the direct and
indirect reflected ray geometry.
1
The equations that follow from this model are a first-order approximation to a more complicated possible
derivation. This approximation is considered acceptable for the aviation-related analyses in this report.

10
Figure F-1: Level-Terrain Two-Ray Path Model Geometry
With the aircraft antenna height denoted by HA and the interference source antenna height by HE
the direct and reflected ray path lengths are defined respectively as: ( ) ( ) ( )
2 2
DIR A E
R r H H r
= − +
and ( ) ( ) ( )
2 2
REFL A E
R r H H r
= + + . The reflected ray relative phase lag is determined by
( )
2
( ) ( ) ( )
( ) REFL DIR
c c
r R r R r
f
 
π
φ
= −
 
λ
 
, where λc is the free-space wavelength at the receiver center
frequency, fc. The grazing angle, θGR(r), of the reflected ray with the concrete reflecting surface
is given as
1
( ) sin (( ) / ( ))
GR A E REFL
r H H R r
−
θ = + . Two electrical parameters of concrete (relative
dielectric constant, εr = 7.0, conductivity, σcc = 0.15 S/m) are used to form a constituent ratio
parameter,
0
( )
2
cc
c
c
x f
f
σ
=
π ε
, where ε0 is the free space permittivity. With the ratio parameter x
defined, the complex reflection coefficient for vertical polarized waves, ρv(r), is given as:
2
2
( ( ))sin( ( )) ( ( )) cos ( ( ))
( , )
( ( ))sin( ( )) ( ( )) cos ( ( ))
r c GR r c GR
v c
r c GR r c GR
i x f r i x f r
r f
i x f r i x f r
ε − ⋅ θ − ε − ⋅ − θ
ρ =
ε − ⋅ θ + ε − ⋅ − θ
;
where the imaginary constant, 1
i= − . The complex multipath field factor at the receive antenna
is then given by
( )
( )
( )
( , ) 1 ( , ) i r
DIR
v c v c
REFL
R r
r f r f e
R r
− φ
 
Ρ =
+ ρ ⋅
 
 
. With these definitions the two-
ray isotropic median path spreading factor (reciprocal of path loss) is written (algebraic terms)
as:
HA
HE
HA - HE
r
RDIR
θGR
RREF
θAE
θEA

11
2
2
( , )
( )
( , )
4 ( )
v c
c c
Ray c
DIR
r f
f
PF r f
R r
 Ρ 
λ
= ⋅
 
π
 
for 0 ≤ r < r1.
(F-1)
The aircraft and source antenna gains for the direct ray path can be computed from the direct ray
elevation angle,
1
( ) sin (( ) / ( ))
AE A E DIR
r H H R r
−
θ =
− − . Note that θEA = -θAE. The break point r1 is
described in Section F.2.2.2. Figure F-2 shows a typical two-ray model path spreading factor
plotted in dB as a function of the range r in meters for fc.= 1575.42 MHz
Figure F-2: Example Level Terrain Two-Ray Isotropic Median Path Spreading Factor
F.2.1.1.2 Sloping Terrain Two-Ray Isotropic Path Model
A more general purpose two-ray isotropic path model for use in some scenarios is one with
sloping terrain between emitter and aircraft. The simpler two-ray level terrain model discussed in
Section F.2.1.1.1 must be adjusted as shown in Figure F-3. The sloping terrain grazing angle,
θGRS(r), as a function of the lateral separation radius, r, is derived in terms of known path
parameters: RDIR, HE (as defined in level terrain case), terrain slope angle θS, and the aircraft
height above the local terrain, hA. While the triangles formed by the antennas, ray paths, and
level terrain in Figure B-1 are right similar triangles, the associated triangles in Figure B-3 on
sloping terrain are not right similar. Thus derivation of θGRS(r) will involve use of the Law of
Sines.
2 5 10 20 50 100
rm
eters

75

70

65
PF2Rayr,fcdB

12
Figure F-3: Sloping Terrain Two-Ray Path Model Geometry
The derivation of begins by applying the Pythagorean Theorem to four ground segments,
r2
+ hS
2
= (Rs1 + Rs2)2
.
Using the Law of Sines, the two sloping ground segments Rs1 and Rs2 can be defined in terms of
ground slope and grazing angles and the closer of the emitter or aircraft height;
Rs1 = HE sin(π/2 – (θGRS(r) - θS))/ sin(θGRS(r)) and
Rs2 = hA sin(π/2 – (θGRS(r) + θS) ) / sin(θGRS(r)).
With a = π/2 and b = (θGRS(r) ± θS) used in the identity 𝑠𝑠𝑠𝑠𝑠𝑠(𝑎𝑎 − 𝑏𝑏) = 𝑠𝑠𝑠𝑠𝑠𝑠(𝑎𝑎)𝑐𝑐𝑐𝑐𝑐𝑐(𝑏𝑏) −
𝑐𝑐𝑐𝑐𝑐𝑐(𝑎𝑎)𝑠𝑠𝑠𝑠𝑠𝑠(𝑏𝑏), the relationships for Rs1 and Rs2 simplify to:
Rs1 = HE cos(θGRS(r) - θS) / sin(θGRS(r)) and
Rs2 = hA cos(θGRS(r) + θS) / sin(θGRS(r)).
Applying the identity, 𝑐𝑐𝑐𝑐𝑐𝑐(𝑎𝑎 − 𝑏𝑏) = 𝑐𝑐𝑐𝑐𝑐𝑐(𝑎𝑎)𝑐𝑐𝑐𝑐𝑐𝑐(𝑏𝑏) + 𝑠𝑠𝑠𝑠𝑠𝑠(𝑎𝑎)𝑠𝑠𝑠𝑠𝑠𝑠(𝑏𝑏), to the expression for Rs1
and the identity, 𝑐𝑐𝑐𝑐𝑐𝑐(𝑎𝑎 + 𝑏𝑏) = 𝑐𝑐𝑐𝑐𝑐𝑐(𝑎𝑎)𝑐𝑐𝑐𝑐𝑐𝑐(𝑏𝑏) − 𝑠𝑠𝑠𝑠𝑠𝑠(𝑎𝑎)𝑠𝑠𝑠𝑠𝑠𝑠(𝑏𝑏), to the expression for Rs2, then
substituting both results back into the original Pythagorean Theorem expression of the four
ground segments, and taking the square root of both sides of the expression yields:
(r2
+ hS
2
)1/2
= {HE [cos(θGRS(r))cos(θS) + sin(θGRS(r))sin(θS)]
+ hA [cos(θGRS(r))cos(θS)- sin(θGRS(r))sin(θS)]} / sin(θGRS(r)).

13
Finally, applying the trigonometric relationship hs = r[tan(θS)] for the slope height above level
terrain assumption at the aircraft nadir, and the identity, cos(a)/sin(a) = 1/tan(a), and performing
algebraic simplifications the resulting expression is:
( ) ( )
( ) ( ) ( )
1
2
cos
( ) tan
1 tan sin
E A S
GRS
S E A S
H h
r
r H h
−
 
+ θ
 
θ =
 
+ θ − − θ
 
.
The direct path length, RDIR,S(r), needed for the sloping terrain two-ray path loss is determined by
first noting that the aircraft antenna height above level terrain in Figure F-3 is given by
HA(r) = hA + r tan(θS). Then the level terrain expression, ( ) ( )
2 2
DIR A E
R r H H r
= − + , after
substitution becomes, on the sloping terrain, ( ) ( )
2 2
, tan( )
DIR S A S E
R r h r H r
θ
= + ⋅ − + . The reflected
ray path length, RREF,S(r), is determined from Law of Sines relations to be
RREF,S(r) = (HE + hA)cos(θS)/sin(θGRS(r)). With these revised sloping terrain parameter
definitions, the remaining two-ray path loss factors can be determined from the same expression
forms as in Section F.2.1.1.1.
Once θGRS(r) is determined, four elevation angles, θEAS_DIR, θEAS_REF, θAES_DIR, θAES_REF, can be
derived for use to compute the antenna gains in the scenario. By definition, θS ≥ 0, θGRS > 0,
θEAS_REF < 0, and θAES_REF < 0. It is also true that θAES_DIR = -θEAS_DIR. However, the direct ray
elevation angles θEAS_DIR and θAES_DIR can be either positive or negative depending on the relative
heights of HE vs. (hA + hS). The emitter direct path elevation angle, θEAS_DIR(r), for sloping
terrain is given by: θEAS_DIR(r) = -tan-1
{ [ (HE - hA) – r tan(θS)] / r} for HE > ( hA + hS) and r ≤ r1.
Expressions for the two reflected path elevation angles are θEAS_REF(r) = -(θGRS(r) - θS) and
θAES_REF(r) = -(θGRS(r) + θS).
F.2.1.1.3 Two-Ray Path Model Aspects for Directive, Dual-Polarization Source Antennas
Although RFI sources in some scenarios can be modelled with vertically polarized emissions,
other scenarios call for the use of relatively directive base station source antennas that transmit a
±45° cross-polarized signal. This type of signal is modelled for aviation GPS RFI analyses as a
dual vertically- and horizontally-polarized signal. Each polarization component is assumed to
contain half the total emission power. This signal type causes increased complexity especially in
the two-ray path loss segment. Compared to the formulation in Section F.2.1.1.1, one extra
parameter needed in the two-ray zone is the horizontal-polarized reflection coefficient expressed
as
2
2
sin( ( )) ( ( )) cos ( ( ))
( , )
sin( ( )) ( ( )) cos ( ( ))
GR r c GR
h c
GR r c GR
r i x f r
r f
r i x f r
θ − ε − ⋅ − θ
ρ =
θ + ε − ⋅ − θ
.
Without accounting yet for any source antenna directivity, the associated horizontal polarized

14
multi-path field factor is expressed in similar form to Section F.2.1.1.1 as
( )
( )
( )
( , ) 1 ( , ) i r
DIR
h c h c
REFL
R r
r f r f e
R r
− φ
 
Ρ =
+ ρ ⋅
 
 
; where the direct and reflected
ray path lengths, RDIR(r) and RREFL(r), and the reflected ray path relative phase lag, φ(r), are the
same as for vertical polarization.
To properly account for the source and receive antenna pattern effects on the reflected ray
relative strength compared to the direct ray at the receive antenna, the ray angles need to be
determined. The ray angle expressions in Section F.2.1.1.2 above can be adapted as needed for
flat or sloping ground (flat ground has θS=0). The source side ray angles, once determined, are
then transformed as in Appendix G into the tilted beam coordinates and used to determine the
source antenna power gain in the direct and reflected ray directions. The source antenna power
gain in any direction is assumed to be independent of polarization. The receive side ray angles
are used with the receive antenna equations in Section 4.1.2 to determine the receive antenna
power gain to both direct and reflected vertical and horizontal polarization ray components.
The source transmit direct- and reflected-ray power gain values, Gtd(r) and Gtr(r), are then
formed into a transmit voltage gain ratio, ( ) ( )
( )
Gtr r
Gt
VG rd r
d
t
r
= (same value for both v-
pol. and h-pol.). Similarly the receive direct- and reflected-ray vertical- and horizontal-
polarized power gain values, Grdv(r), Grrv(r), Grdh(r),and Grrh(r) are formed into receive voltage
gain ratios ( ) ( )
( )
v
v
v
VGrd
Grr r
Grd
r
r
= and ( ) ( )
( )
h
h
h
VGrd
Grr r
Grd
r
r
= . These voltage gain
ratios are used to modify the multi-path field factor and yield the directive multi-path factors:
( )
( )
( )
( , ) 1 ( ) ( ) ( , ) i r
DIR
h c h h c
REFL
R r
d r f VGtrd r VGrd r r f e
R r
− φ
 
Ρ =
+ ⋅ ⋅ ρ ⋅
 
 
and
( )
( )
( )
( , ) 1 ( ) ( ) ( , ) i r
DIR
v c v v c
REFL
R r
d r f VGtrd r VGrd r r f e
R r
− φ
 
Ρ =
+ ⋅ ⋅ ρ ⋅
 
 
.
Since the voltage gain ratios are less than unity, their effect in the above expressions is to
diminish the magnitude of the fluctuating component in the multi-path field factor at a given
radius compared to the initial value without the directivity correction.
When these modified multi-path field factor expressions are substituted into the basic two-ray
path factor Equation (F-1), the resulting vertical and horizontal polarization path factors are:
2
2 ,
( , )
( )
( , )
4 ( )
v c
c c
Ray v c
DIR
d r f
f
PF r f
R r
 Ρ 
λ
= ⋅
 
π
 
and
2
2 ,
( , )
( )
( , )
4 ( )
h c
c c
Ray h c
DIR
d r f
f
PF r f
R r
 Ρ 
λ
= ⋅
 
π
 
.
When the effects of a directive dual-polarized base station source antenna are incorporated, as
discussed above, an additional complication occurs in the determination of the outer radius limit,
r1, for the two-ray path model. Investigation of the path factor expressions indicates that at any

15
particular given radius the horizontal and vertical path factors may have different values.
However, there are certain radius values at which the path factors are equal, a feature which is
necessary for the starting radius of the mid-range path model zone.
For sources emitting only vertically polarized signals, the r1 value is the radius at which the
complex vertical polarization reflection coefficient, ρv(r, fC), is purely imaginary (phase of ρv
= -90°) very near the minimum magnitude. At this radius, the grazing angle for the reflecting
surface is at a critical value, θCRIT, that depends on only the surface electrical parameters. With
the critical grazing angle known along with antenna heights, the expression for grazing angle in
Section F.2.1.1.2 can be inverted to solve for the associated vertical polarization radius, r1v. The
strategy for the dual polarization radius, r1d is to determine from the path factor equations, the
first radius beyond r1v at which the vertical and horizontal polarization path factors are equal
(usually a few centimeters beyond r1v). The two-ray path factor value at r1d is used in the formula
for the mid-range path factor.
F.2.1.2 Medium Range Path Segment Erceg-Greenstein Isotropic Model
Median path propagation models were initially developed for cellular mobile phones at lateral
separation ranges beyond 1 km and out to 20 km. This left a need for some type of model to
cover ranges from 100 meters to 1 km. In the mid-1990's this need was met by researchers from
ATT Laboratories who developed a mid-range model that came to be known as the Erceg-
Greenstein model [4]. Their model pertains to ranges from 100 meters to 8 km with base station
antenna heights between 10 and 80 meters and a mobile unit antenna height of 2 meters. The
model was developed for suburban environments using three different terrain categories to make
it more flexible. An operating frequency of 1.9 GHz was used in developing the model although
extrapolations to other frequencies are possible. In the Erceg-Greenstein model the median path
loss is proportional to ( / )
o
r r γ
where ro is usually taken to be 100 meters. The loss exponent γ
changes with terrain type but is greater than 2 meaning that the path loss is more severe than that
associated with a free space model. The classic Erceg-Greenstein model has the actual path loss
increasing slightly as the range increases, however the modified model used in this analysis
holds this constant with range.
The modified Erceg-Greenstein model has a median isotropic path spreading factor (in algebraic
terms) given by:
2
,mod
0
( , ) 1/ ( )
EG
A
A
c
a b H
H
EG c c
r
PF r f A f
r
− ⋅ +
 
 
 
= ⋅ 
 
 
 
 
, for r1 ≤ r < r2
(F-2)
In Equation (F-2), A(fc) = 4πr0/λc(fc) with r0 = 100 m. Values for constants a (3.6), b (0.005), and
cEG (20.0) are chosen to represent the area in the vicinity of an airport (terrain model C (flat, light
tree cover)). The break points r1 and r2 are described in Section F.2.2.2. Figure F-4 shows a

16
sample median path spreading factor produced by the Erceg-Greenstein model at fc = 1575.42
MHz.
Figure F-4: Example Erceg-Greenstein Isotropic Median Path Spreading Factor
F.2.1.3 Medium Range Path Segment Exponential Fit Isotropic Model
For cases when HA > 80 m (e.g., the FAF Waypoint) or HE > 2 m, an exponential, constant slope
blending function is used. The function’s constant exponent and intercept point are chosen to
achieve continuity at either end with the short- and long-range path loss functions given
appropriate segment breakpoints, r1 and r2. For handset interference sources, the exponential
constant slope path loss model is given by:
1 ( )
exp 1 2 1 1
( , ) ( , )( / ) c
f
fit c Ray c
PF r f PF r f r r −Γ
= (F-3)
The exponent is expressed as:
1 1 2 10 2 10 2 1 10 2 1
( , , ) [ ( ) ( /1000) ( ( , ))]/ ( / )
c c c Ray c
r r f f log r log PF r f log r r
α β
Γ = + ⋅ +
where r2 < 20,000 m, and α(fc), and βc, are parameters associated with the Hata-Okumura model
discussed in Section F.2.1.4.
For base station interference sources there is a single exponential constant slope model given in
similar form but with different parameter values by
1 1 2
( , , )
exp 1 2 1 1
( , ) ( , )( / ) c
r r f
fit c Ray c
PF r f PF r f r r −Γ
= (F-4)
where Γ1(r1,r2, fc) has the same form as in Equation (F-3).
F.2.1.4 Long Range Segment Path Isotropic Model
500 1000 2000 5000
rm
eters

140

130

120

110

100
PFE G m
odr,fcdB

17
For propagation over “flat earth” terrain from RFI sources at distances beyond 1 km from the
aircraft antenna, the Hata-Okumura median path propagation model is used. It was originally
developed for predicting UHF cellular mobile telephone propagation at distances beyond 1 km
from the base station for three terrain types: urban areas, suburban areas and open areas. In this
Report the suburban terrain parameters are used to represent the environment around a large
metropolitan airport. As with the Erceg-Greenstein model, the Hata-Okumura path median
spreading factor is inversely proportional to (r/ro)γ but in this case ro is set to 1000 meters. The
exponent γ varies with antenna height as well as range and for distances beyond 20 km it also
varies with operating frequency. With the ITU-R extension to the original model, it is usable to
ranges of 100 km. Figure F-5 shows the median path spreading factor generated by the Hata-
Okumura model for an operating frequency fc = 1575.42 MHz.
Figure F-5: Example Hata-Okumura Median Isotropic Path Spreading Factor
The Hata suburban median isotropic path spreading factor (in algebraic terms) is given by:
( )
( )
( , , )
( ( )/10 ( ) (log /1000 ) )
( , ) 10
F r f H
c A
sp c c
K r f r
HS c
PF r f
− α +β ⋅
= (F-5)
for r2 ≤ r ≤ R0 as appropriate where ( )
sp
K r (in dB) denotes a range-dependent slope correction
factor (0 dB for “flat-earth”),
( ) ( ) ( )
2 2
69.12 26.16 log 2 log 13.82 log 3.2 log 11.75
28
( ) 0.1
fc
f H H
c E
A
fc + ⋅ − ⋅ − ⋅ − ⋅
 
 
α =  
 
 
 
,
( )
0.1 44.9 6.55 log
C A
H
β = ⋅ − ⋅
 
  and F(r,fc,HA)= 1, r ≤ 20 km, or:
( ) ( )
( )
( )
3
4 4 0.8
10
6 2
1.07 10
, , 1 0.14 1.87 10 (log / 2 10 )
1 7 10
A
c A c
A
H
F r f H f r
H
−
−
−
 
⋅
 
=
+ + ⋅ + ⋅
+ ⋅
 
 
for r>20km.
10000 20000 30000
15000
rm
eters

150

145

140

135

130
PFH Sr,fcdB

18
For the certified aviation equipment analyses in this report, the Hata “base station” antenna
height, HA, in the long range model segment is assumed to be the greater of the aircraft or
interference source antenna height. The Hata “mobile station” height, HE, is the lesser of the two
antenna heights. In addition, the propagation is assumed to be reciprocal.
When applying the Hata suburban median isotropic spreading factor to certain discrete base
station scenarios on irregular, sloping terrain, it may be necessary to apply a non-zero slope
correction factor ( )
sp
K r to the model. This correction factor, derived from the work of Okumura
[2], was chosen as the principal factor and other smaller ones were considered negligible. These
terrain slopes are averages computed over a significant distance (Okumura suggests distances of
5-10 km), The correction as given in [2] shows that positive (upward, away from the aircraft)
slopes reduce the path loss while negative (downward) slopes yield increased path loss. Steep
slopes produce increased correction whether they are positive or negative. The correction is
given in dB as a function of slope angle in milliradians and distance in kilometers. Correction
factor data points presented graphically in [ [2], (Figure 34)] have been extracted and fit with an
equation for ease of use. The Okumura data points and fit equations for several distances (10, 30
, 60 km) are plotted in Figure F-6. The fit equation Ksp(θ, d) in dB as a function of slope angle θ
(milliradians) and distance d (m) is:
2
1 2 3 4 5
2 2 2
1 2 3 4 5 6 7
( ( /1000)) ( ( /1000)) , 0
( , )
( ( /1000) ( /1000) ) ( ( /1000) ( /1000) ) ,0
a a a d a a d
Ksp d
b b b d b d b b d b d
θ θ θ
θ
θ θ θ
 
+ + − + <
 
=  
+ + + − + + ≤
 
 
where a1 = 0.052857, a2 = -0.105285, a3 = 0.0288714, a4 = 0.0027143, a5 = 7.14265x10-5
,
b1 = 0.018567, b2 = 0.01394, b3 = 0.0193213, b4 = -9.1533x10-5
, b5 = -1.94286x10-3
,
b6 = 4.25715x10-4
, and b7 = -3.14287x10-6
.
Figure F-6: Okumura Slope Correction Factor Data Points and Fit Equations
In applying this factor, it proved useful to divide the terrain into angular sectors about the aircraft
nadir point. These sectors were chosen so as to provide as much terrain uniformity as possible

20 
10 10 20

m
radians

15

10

5
5
10
KspdB 60
km
30
km
10
km
30
km
10
km

19
when traversing outward from aircraft nadir along any given radial line. Within a given sector,
an average ground height was determined extending to some appropriate radial distance from
aircraft nadir followed by a slope determination, if one exists beyond such distance.
F.2.1.5 Free Space Path Isotropic Model
For aircraft antenna heights generally greater than 550 meters and certain special cases below
550 m, there is predominately a clear line of sight to all RFI sources other than ground based
mobile sources whose interference impact has been shown to be negligible at these altitudes from
previous analysis [5]. In such cases, a free space path propagation model is most appropriate.
Free space propagation is the most basic model in which the path isotropic spreading factor is
proportional to the inverse square of the separation range. This is a deterministic model with no
associated probability distribution. Hence the path spreading factor can be exactly determined
once the range between GPS antenna and interfering source is known. The free space isotropic
spreading factor is given by:
2
( , ) ( / 4 ( ) )
FSP c DIR c
PF r f c R r f
π
= (F-6)
where, as previously defined, ( ) ( ) ( )
2 2
DIR A E
R r H H r
= − + and c is the speed of light in vacuum.
F.2.2 Single Path Probabilistic Propagation Model
F.2.2.1 Single Path Probabilistic Propagation Model Basic Aspects
The probabilistic propagation channel is typically modeled as a product of a slow fading process
(log-normal power distribution) and a fast fading process [3]. The fast fading portion may be
either "flat" (maximum delay spread, τds << 1/W, where W=interference signal base-bandwidth)
or it may be frequency selective. The simplest case to model is the case where the fast fading is
flat. In that case, the received interference from a single emitter, γ(t), may be written as
 
{ }
( ) Re ( ) ( ) ( )
E
t t t s t
γ α β
= where Re{ } denotes the real part, while 
α , 
β , and sE are
independent complex random processes. The term ( )
t
α denotes the fast fading process, ( )
t
β the
slow fading component, and ( )
E
s t the portion of the interfering emitter signal arriving at the GPS
receiver antenna at center frequency fc, (sE(t) = u(t)exp(i2πfct), where u(t) is complex).
In the propagation model, 
α has independent normally distributed real and imaginary parts each
with mean values that may be different from zero to account for the presence of a strong line-of-
sight component at shorter ranges. Thus,  2
( )
t
α , has a non-central Chi-squared distribution with
two degrees of freedom. Although
2
( )
E
s t is a random process, to avoid dealing with those
statistics the instantaneous power associated with the single source emission (unwanted emission
in this analysis) is deemed to be constant as in most of the literature on path loss models. In the

20
model used for this analysis, power was assumed to be 2 o
PW watts; i.e., an rms power spectral
density of P0 watts/Hz over a signal base-bandwidth, W. Under these assumptions, the random
process  2
( ) ( )
E
t s t
β is log-normally distributed and the received single emitter interference power
2
( )
t
γ becomes the product of two random variables, one having a non-central Chi-squared
distribution while the other is log-normally distributed.
For a scenario with frequency selective fast fading, it is shown [6] that γ can be approximated as
the sum of several independent flat fading processes. Thus, γ(t) is written as
 
1
0
( ) ( ( ) ( )) ( )
L
l E l
l
t t s t t
γ α τ β
−
=
= −
∑ ; where the channel parameter, L, is the number of resolvable fast
fading paths and the l
τ are chosen such that [ ( ) ( )] 0
E i E k
E s t s t
τ τ
− − =
for k l
≠ . Then the
instantaneous power,
2
( )
t
γ , becomes
1
2 2 2 2
0
( ) ( ( ) ( ) ) ( )
L
l E l
l
t t s t t
γ α τ β
−
=
= −
∑ . .
Using the same assumption as above regarding the single emitter power, the received single
emitter power density may be written as:
 
1 2 2
2
0
( ) ( ( ) ) ( )
SlowFading
L
l o
density
l
FastFading
t t t P
γ α β
−
=
= ∑





.
This form for the received single emitter power density is the most flexible and is the form
adopted in the present model. The slow fading portion is log-normally distributed with
parameters µ and σ while the fast fading process has a non-central Chi-squared distribution with
2L degrees of freedom and parameters ρo and ψo.
The propagation model assumes each of these parameters to be range dependent hence we may
express the received single emitter power spectral density as:
( ) ( ) ( )
E
P r r r
=
Σ Ω ,
where Σ denotes the fast fading component and Ω the slow fading. Note all time dependence has
been dropped as all of the constituent processes are assumed to be stationary.
F.2.2.2 Median Path Segment Propagation Parameters and Segment Break Points
The propagation environment is modeled probabilistically with a range and height-dependent
median path loss to an individual RFI source. Based on a flat-earth assumption, a continuous
median path spreading factor function, PF(r,fc), is generated by combining three range segment
models taken in part from mobile radiotelephone propagation studies. The long range segment
model (r typically greater than 1 km) is based on the Hata-Okumura model [ [1], (suburban

21
case)]. Two options are used for the mid-range segment model depending on aircraft antenna
height: Erceg/Greenstein (HA ≤ 80 m), and an exponential constant slope blend models for HA >
80 m. The short range segment model (ranges less than a few hundred meters) is the classic two-
ray with concrete as the single reflecting surface. Reciprocity is assumed to hold as it relates to
the probabilistic modeling of the propagation channel. The median path isotropic spreading
factor is then expressed for handset interference sources as:
2 1
,mod exp 1 1 2
2
( , ), 0
( , ) ( , ) ( , ),
( , ),
Ray c
c EG c fit c
HS c o
PF r f r r
PF r f PF r f or PF r f r r r
PF r f r r R
≤ <
 
 
≤ <
 
 
≤ ≤
 
(F-7)
while, for base station interference, sources:
2 1
exp 1 1 2
2
( , ), 0
( , ) ( , ),
( , ),
Ray c
c fit c
HS c o
PF r f r r
PF r f PF r f r r r
PF r f r r R
≤ <
 
 
≤ <
 
 
≤ ≤
 
(F-8)
For handset interference sources, the following set of guidelines was used for the isotropic
median path loss break points.
• At short ranges a two-ray median path propagation model is used up to the range r1 where the
two ray and Erceg-Greenstein and/or exponential fit models join. This break point varies with
aircraft antenna height and is computed so as to produce a continuous path spreading factor
function. When the Erceg-Greenstein model is used, this break point is usually near 100
meters. When the exponential fit model is used, as in the FAF scenario, this break point is
near where the two ray vertically polarized component reflection coefficient is at minimum
magnitude.
• The second segment uses the Erceg-Greenstein model, if applicable, up to the point r2 where
it intersects the Hata-Okumura median path propagation model. When the exponential fit
model is used, to preserve some consistency with the lower height cases, the mid-range
segment inner radius, r1, is set to give a two-ray segment grazing angle, θGR(r) such that
tan(θGR(r1)) ≅ 0.5. Some slight adjustment of the r1 value may be done, if needed, to reduce
the spreading factor slope change at the junction with the two-ray segment. In a similar way,
the outer radius, r2, is set to give an elevation angle from the source to the receive antenna,
ε(r), such that tan(ε(R2)) = 1/14 (i.e.; ε(r2) ≅ 4°). An exponential fit mid-range segment is also
used for very low aircraft antenna heights (e.g. < 10 m). In these instances, r1 is set for the
radius at which the two ray vertically polarized component reflection coefficient is at
minimum magnitude. The outer radius, r2, is set at 1 km (the minimum applicable radius for
the Hata model).
• Beyond r2 the Hata-Okumura suburban median path propagation model is used.
• The radio horizon is given by ( )
0 4124.12 A E
R H H
= + (all dimensions in meters).

22
For base station interference sources, two basic breakpoints were defined as follows.
• At short ranges a two-ray median path propagation model is used up to the radius r1 where
the vertically polarized component reflection coefficient is at minimum magnitude. This
break point varies with aircraft antenna height.
• Beyond r1, the median path spreading factor is modelled as an exponential constant-slope
blending function out to radius, r2 (details in Section F.2.1.3), where it intersects the Hata-
Okumura long range segment.
• Beyond r2, the Hata-Okumura suburban median path propagation model is used.
F.2.2.3 Single Path Probabilistic Propagation Slow Fading Parameters, µ and σ
With the median path spreading factor, PF(r,fc), defined for any individual source, the single-
path slow fading parameter, µ(r,fc), can be written in general as:
( , ) [ ( ( ), ) ( , ) ( ( ))]
c xmit xmit elev az c rcv elev
r f ln P G r PF r f G r
µ θ φ ζ
= ⋅ ⋅ ⋅
This expression can be rewritten and further simplified to:
0
( , ) ln( ) ( , )
c c
r f P r f
µ µ′
= + (F-9)
where ( , ) [ ( ( ), ) ( , ) ( ( ))]
c norm elev az c rcv elev
r f ln G r PF r f G r
µ θ φ ζ
′
= ⋅ ⋅ is a unitless mean fading factor,
ln() denotes natural logarithm and P0 (Pxmit η G0) is the RFI source effective isotropic radiated
power (EIRP) or power spectra density at the transmit beam center (Appendix G.1). The antenna
gain factors are: Gxmit( ) the interfering emitter antenna gain, θelev(r) the transmit elevation angle,
and φaz the azimuth angle toward the receive antenna, Gnorm() is the normalized transmit gain
(Gxmt( )/η G0), Grcv( ) the receive antenna gain, and ( )
elev r
ζ the receive elevation angle toward
the interfering emitter antenna. Note that Gnorm( ) is set to unity independent of angle for handset
emitters.
The single-path slow fading parameter, σ, is generally range dependent and hence is written as
σ(r). Its range dependence varies with the interference scenario, i.e., whether the interference is
from a base station source or from a handset source. For a handset source, a linear transition
region is used between the very short range condition and the onset of significant scattering
(chosen to be the mid-range segment inner radius) [5]. The transition region starting point is
chosen as the radius, rs, such that the elevation angle from the source to the receive antenna, ε(rs)
is 45° (i.e.; rs = HA-HE)2
. The transition end point is the radius, r1, previously defined in Section
B.2.2.2 for handset sources. For convenience, the single-path standard deviation, σ(r), is
2
For handset scenarios when the aircraft antenna height is very low (e.g. < 10 m), the transition region start point
radius, rs is set equal to r1 and the end point radius is r2.

23
described in decibel terms. Similar to the strategy used in [B-5], the short-range and full-
scattering σ(r) limits used in this analysis are 0.5 dB and 6.4 dB3
, respectively. Thus:
( )
( )
( ) 1
1
1
0.5, 0
0.5 5.9 ,
6.4,
s
s
dB s
s
r r
r r
r r r r
r r
r r
σ
≤ ≤
 
 
−
 
= + < ≤
 
−
 
 
>
 
(F-10)
Note that ( ) ( (10) /10) ( )
dB
r Ln r
σ σ
= .
For base station sources the method used in [5] is modified somewhat in this analysis. Instead of
a stepped or polynomial function for the radial variation, a linear function for σdB(r) similar to
that used for handsets is used. A standard deviation for "light shadowing" is given as 0.5 dB,
which is similar to the handset scenario is assumed for short distances (< r1), in agreement with
Loo's result [7]. Beyond the r1 breakpoint, σdB(r) rises linearly in r up to 6.4 dB at the r2
breakpoint (start of Hata Okumura path segment). (See Section F.2.2.2 for base station sources).
Thus:
( )
( )
( )
1
1
1 2
2 1
2
0.5, 0
0.5 5.9 ,
6.4,
dB
r r
r r
r r r r
r r
r r
σ
≤ ≤
 
 
−
 
= + < ≤
 
−
 
 
>
 
(F-11)
F.2.2.4 Single Path Probabilistic Propagation Fast-Fading Parameters, L, ψ0, ρ0
For handset interference sources, the unitless diffuse scattering parameter4
, ψ0(r), is specified in
terms of a product with channel parameter, L, as:
3
The 6.4 dB standard deviation value used in the Log-Normal slow fading component of the certified avionics
analysis model comes from measurements made around London by M.F. Ibrahim and J.D. Parsons [3]. This 6.4 dB
was applicable for London type terrain and measured at a frequency of 900 MHz. As shown in [2], the standard
deviation does not change much with frequency, so a 6.4 dB standard deviation was also assumed for the certified
avionics analysis around 1531 MHz. In reality the standard deviation may be around 0.5 dB greater at 1531 MHz
than at 900 MHz.
4
For handset scenarios when the aircraft antenna height is very low (e.g. < 10 m), the fast fading parameters
transition region start point radius, rs is set equal to r1 and the end point radius is r2.

24
( )
( )
( )
0 1
1
1
0.1, 0
2 0.1 0.9 ,
1.0,
s
s
s
s
r r
r r
L r r r r
r r
r r
ψ
≤ ≤
 
 
−
 
= + < ≤
 
−
 
 
>
 
(F-12)
Note in Equation (F-12) that the composite parameter 2Lψ0(r) is non-zero at short ranges (< rs).
This aspect provides consistency with the assumption that σ(r) is also non-zero over the same
ranges. This result associates the small σ(r) value with a small amount of diffuse scattering in the
propagation loss at short ranges. For this analysis DS
L W
τ
=  
  where the operator  
  implies
rounding up the operand to the next integer, τDS, is the spread in channel delay time, and W is the
channel baseband bandwidth. Thus L denotes the number of resolvable paths associated with the
fast fading component (see Section F.2.2.1).
In a similar manner then have for base station interference sources:
( )
( )
( )
1
1
0 1 2
2 1
2
0.1, 0
2 0.1 0.9 ,
1.0,
r r
r r
L r r r r
r r
r r
ψ
≤ ≤
 
 
−
 
= + < ≤
 
−
 
 
>
 
(F-13)
For handset sources the unitless line-of-sight parameter, ρ0(r), is specified as:
( )
( )
( )
2
0 1
1
1
1.0, 0
1.0 ,
0,
s
s
s
s
r r
r r
r r r r
r r
r r
ρ
≤ ≤
 
 
−
 
= − < ≤
 
−
 
 
>
 
(F-14)
Again, for the base station scenario using the break point definitions given in Section F.2.2.2:

25
( )
( )
( )
1
1
2
0 1 2
2 1
2
1.0, 0
1.0 ,
0,
r r
r r
r r r r
r r
r r
ρ
≤ ≤
 
 
−
 
= − < ≤
 
−
 
 
>
 
(F-15)
F.3 Aggregate Propagation Effects Computation Methods
Aggregate propagation effects are treated differently depending upon whether the interference
sources are randomly distributed or have discrete known locations. Although there are
exceptions, most handset sources use randomly distributed sources while base station sources
have known locations. For uniformly-distributed, randomly-located interference sources, it is
shown ( [8], [9]) that once the single path interference characteristics have been determined, it is
possible to determine the mean power, standard deviation, and cumulative probability
distribution associated with aggregate received interference power. The same holds true for
sources that have discrete known locations although the method used to determine mean power
and cumulative probability distribution is different.
F.3.1 Randomly-Located Source Statistics
F.3.1.1 Randomly-Located Source Mean Aggregate Interference Power
As noted, for uniformly-distributed, randomly-located interference sources, once the single path
interference characteristics have been determined, it is possible to determine the mean power,
standard deviation, and cumulative probability distribution associated with aggregate received
interference power by the “Analytic Statistical” method. For example, the aggregate mean power
is given by:
2 2
0
(2 ( ) ( ))exp[ ( ) / 2 ( , )] ( )
o
R
I o o o c
P P n L r r r r f f r dr
ψ ρ σ µ′
= + +
∫ (F-16)
where n is the mean number of active emitters (Poisson-distributed) within the radio horizon
radius, R0, and f(r) is the radial location probability density function for uniformly distributed
emitters. (More details in [9]). The expression, 2
(2 ( ) ( ))
o o
L r r
ψ ρ
+ , is the first moment (mean) of
the non-central Chi-squared fast-fading distribution and the expression, exp[σ2
(r)/2 + µ’(r, fc)], is
the log-normal slow-fading distribution unitless mean.
F.3.1.2 Randomly-Located Source Aggregate Interference Power Standard Deviation
For randomly-located interference sources, the normalized standard deviation (normalized by the
aggregate mean power) has been determined as in [9] by:

26
2 2 2 2 2
((2 ( ) ( )) 4 ( ) 4 ( ) ( )) [2 ( , ) 2 ( )] ( ) /
0 0
Ro
P n L r r L r r r exp r f r f r dr P
o o o o o
I I
c
σ ψ ρ ψ ψ ρ µ σ
′
=
⋅ + + + +
∫ (F-17)
where the parameters L, µ, σ, ψo, and ρo are as defined previously.
F.3.1.3 Randomly-Located Source Aggregate Interference Power Probability Distribution
Following the methodology described in [9] for randomly-located sources, the characteristic
function associated with the normalized aggregate interference power is expressed as:
( ) exp[ (1 ( ))]
C n
τ φ τ
= − − (F-18)
where ( ) [exp( ( ) / )]
E I
E i P r P
φ τ τ
= , i.e., ( )
φ τ is the characteristic function of the interference from
a single emitter normalized by the mean aggregate interference power.
It is shown in [B-9], after some manipulation it is possible to write:
2
0
( ) (1/ ) ( , , ) ( ) ( )
o
R
y r f r Exp y drdy
φ τ π τ
∞
−∞
= Φ −
∫ ∫ ; where
2
( , , ) [1 / (1 2 ( ) ( , ) exp( 2 ( ) )) ] exp[ ( , ) exp( 2 ( ) ) ( ) / (1 2 ( ) ( , ) exp( 2 ( ) ))]
c c c
L
y r i r r f r y i r f r y r i r r f r y
o o o
τ τψ σ τ σ ρ τψ σ
Φ = − Γ Γ − Γ
and ( , ) ( ( ), ) ( , ) ( ( )) /
c o norm elev az c rcv elev I
r f P G r PF r f G r P
θ φ ζ
Γ =⋅ ⋅ ⋅
Then as in [10], the K-point Hermite-Gauss quadrature method with weights k
w and nodes k
y
may be used to write:
1 0
( ) ( / ) ( , , ) ( )
o
R
K
k k
k
w y r f r dr
φ τ π τ
=
≈ Φ
∑ ∫ (F-19)
A value of 65 has been used for K to obtain very good accuracy in Equation (F-19).
Finally using Equation (F-18), we obtain the desired cumulative distribution function for the
random variable I
I P
P / as in [11] by the following:
Pr( / ) 1 (1/ (2 )) lim ((1 exp( )) / ( )) ( ) (1/ 2)exp( )
A
I I
A
A
P P z i z i C d n
π τ τ τ τ
→∞
−
> =− − − + −
∫ (F-20)
for .
0
>
z
In those instances where randomly-located handsets are considered and where there is an already
established level of unwanted interference from existing baseline sources, the procedure below
may be used to determine the cumulative distribution function for the total aggregate
interference.

27
Let 1 1
/
I I
P P denote the normalized aggregate interference from existing baseline sources and
2 2
/
I I
P P that from the new handset sources. Also let 1 2
n n n
= + where 1
n and 2
n are the mean
number of active emitters within the radio horizon associated with existing baseline sources and
new handsets, respectively. The mean number of active emitters may be computed as described
in [9] using parameters applicable to either existing sources or new handsets. The characteristic
function 1( )
C τ associated with existing baseline sources is then given by:
1 1 2
1 1 1
( ) [ (1 ( / ( ))]
I I I
C Exp n P P P
τ φ τ
= − − +
where 1( )
φ • is the normalized single emitter interference characteristic function determined using
the methodology described above with the parameters provided in [9]. The function 1( )
φ • must
be computed for whatever flight scenario is applicable (see, for example, [9]).
Similarly,
2 1 2
2 2 2
( ) [ (1 ( / ( ))]
I I I
C Exp n P P P
τ φ τ
= − − +
where 2 ( )
φ • is determined using parameters applicable to the new handsets under consideration.
The characteristic function ( )
C τ associated with the total normalized aggregate interference,
1 2 1 2
( ) / ( )
I I I I
P P P P
+ + , is then given by 1 2
( ) ( ) ( )
C C C
τ τ τ
= , which may be inverted to give the
desired cumulative distribution function using Equation (F-20).
F.3.2 Discretely-Located Source Statistics
F.3.2.1 Discrete Source Mean Aggregate Interference Power
For sources having a known discrete distribution (e.g., the relative location (range) and height
parameters associated with each source are known), two possible approaches may be used to
determine the aggregate interference power, its mean and cumulative distribution function. Since
the single source interference power in this case is still a random variable and is described by the
product of a slow fading process and a fast fading process, similar to the case with randomly-
located sources, the aggregate mean power may be computed as:
2 2
1
(2 ( ) ( ))exp[ ( ) / 2 ( , , , )]
k
n
I o o k o k k k E A c
k
P P L r r r r H H f
ψ ρ σ µ
=
′
= + +
∑ (F-21)
where k
r and k
E
H denote the lateral range to and height of the kth emitter antenna, HA is the
aircraft antenna height, and n is the known number of discrete interference sources (Analytic
Statistical method).

28
Alternatively, it is also possible to use a Monte Carlo simulation to determine the mean
aggregate interference power. In this alternative, the received power from the kth interfering
emitter is modeled as the product of a slow fading process (log-normally distributed) and a fast
fading process having a non-central Chi-Squared distribution. The log-normal component is
generated using the same two parameters, µ and σ as in the random source distribution while the
Chi-Squared process uses the parameters, L, ψo and ρo.
F.3.2.2 Discrete Source Aggregate Interference Power Probability Distribution
F.3.2.2.1 General Discrete Source Aggregate Interference Power Probability Distribution
For discretely distributed interference sources, the characteristic function, ( )
k
φ τ , for the kth
single emitter interference power can be computed as ( ) [exp( ( , ) / ]
k k
k E k E I
E i P r H P
φ τ τ
= . This
characteristic function can be written as:
1
( ) ( / ) ( , , )
M
k m m k
m
w y r
φ τ π τ
=
≈ Φ
∑
where the weights m
w and nodes m
y are as defined previously and k
r is the distance to the kth
interference source. A good value for M would be 65 as in the case of randomly-located,
uniformly distributed interference sources.
The characteristic function for the normalized aggregate interference, C(τ), can then be
determined from:
1
( ) ( )
n
k
k
C τ φ τ
=
= ∏ .
Similar to the uniform randomly distributed interference case, the distribution function is
Pr( / ) 1 (1/ (2 )) lim ((1 exp( )) / ( )) ( )
A
I I
A
A
P P z i z i C d
π τ τ τ τ
→∞
−
> =− − −
∫ (F-22)
F.3.2.2.2 Discrete Source Aggregate Characteristic Function with Sectorization
The previous subsection describes in general the method to compute the aggregate characteristic
function C(τ) from the discretely-located individual base station characteristic functions φk(τ).
However, in the discrete location cases where several angular sectors are defined to enable
terrain modeling, some alterations need to be made. The method described above may be used to
determine the characteristic function Cn(τ) for an individual sector assuming each of these
characteristic functions have been normalized using the mean received aggregate factor for the

29
nth sector FdAGGn (used in place of 𝑃𝑃𝐼𝐼
� in previous section). Then the composite characteristic
function C(τ) is defined in terms of the individual sector characteristic functions Cn(τ) by
∏
=
⋅
=
N
n
AGG
AGGn
n MF
Fd
C
C
1
)
)
((
)
( τ
τ
where n is the individual sector number, N is the total number of sectors, and MFAGG is the
overall mean aggregate factor for all the sectors. Once C(τ) is determined, the desired cumulative
probability distribution function for the location can be computed with the inverse Fourier
transform as in Equation (F-22) above.

APPENDIX G
FAA BASE STATION ANTENNA MODEL

FOR OFFICIAL USE ONLY
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 4
REFERENCES ............................................................................................................................... 5
APPENDIX..................................................................................................................................... 6
Base Station Transmit Antenna Far-Field Gain Pattern Model.............................. 6
G.1 Gain Pattern Model Derivation...................................................................................... 6
G.2 Usage of Antenna Gain Pattern Model .......................................................................... 8

LIST OF FIGURES
Figure A-1: Far Field Pattern Coordinate System .......................................................................... 6
Figure A-2: Base Station Transmit Broad beam Elevation Pattern................................................ 7
Figure A-3: Base Station Transmit Broad beam Azimuth Pattern ................................................. 8

LIST OF TABLES
Table A-1: Base Station Transmit Antenna Parameters ................................................................. 6

REFERENCES

APPENDIX
Base Station Transmit Antenna Far-Field Gain Pattern Model
G.1 Gain Pattern Model Derivation
The base station transmit antenna far field pattern model1
is that associated with a uniformly
illuminated rectangular aperture of width "a" and height "b". The aperture is centered on the "xy"
coordinate center shown in Figure G-1 with the width "a" being co aligned with the "x" axis and
the height "b" with the "y" axis. (Figure G-1 depicts general angular coordinates relative to the
indicated axes.)
Figure G-1: Far Field Pattern Coordinate System
The beam is centered on the "z" axis with a peak gain, elevation beamwidth and azimuth
beamwidth as shown in Table G-1. Table G-1 also lists other important parameters associated
with the far field pattern. Table G-1 values are shown for the “narrow beam” (i.e., the study
described in 5.3.3.3, Results for Inflight Aircraft Ground Based Base Station Cases) and “broad
beam” (i.e., the study described in 5.3.3.6, Results for Inflight Aircraft TAWS/HTAWS and Low
Altitude Pos/Nav) cases.
Table G-1: Base Station Transmit Antenna Parameters
Parameter “Narrow Beam” Value “Broad Beam” Value
a/O 0.8278 0.7722535
b/O 5.7 2.641574
Go 72.5344 31.8649
K 0.681491 0.686578
Peak Gain 16.94 dB 13.4 dB
Elevation Beamwidth 8.8 deg 18.6 deg
1
Pattern based on uniformly illuminated rectangular aperture as described in C. A. Balanis, "Antenna Theory:
Analysis and Design", John Wiley & Sons, 2005 while accounting for a change in coordinate systems from that
used in the book.

Azimuth Beamwidth 64.6 deg 70.0 deg
The equations shown below are used by both the “narrow beam” and “broad beam” models and
the dual use is true for all equations in this appendix unless otherwise indicated.
¿
¾
½
¯
®

elsewhere
and
g
)],
'
(
sin
)
'
(
cos
)
'
(
/[sin
)]
'
(
cos
)
'
(
sin
)
'
(
cos
)
'
(
[sin
}}
{
},
0
{{
'
0
'
,
1
)
'
,
'
( 2
2
2
2
2
4
2
T
T
I
I
T
T
I
S
I
T
I
T
)
'
(
)
'
,
'
(
)]
'
sin(
)
/
(
[
sin
)]
'
cos(
)
'
sin(
)
/
(
[
sin
)
'
,
'
( 2
2
I
J
I
T
T
O
S
T
I
O
S
I
T g
b
c
a
c
h
A variation between the antenna equations of the “narrow beam” and “broad beam” models for
the function, J(I’), is that an exponential coefficient has a value of either a 3 or 4, as shown
below.
Narrow Beam Model:
¿
¾
½
¯
®

d

d
S
I
S
S
I
S
I
I
J
|
'
|
2
)],
2
|
'
(|
3
exp[
2
|
'
|
,
1
)
'
(
Broad Beam Model:
¿
¾
½
¯
®

d

d
S
I
S
S
I
S
I
I
J
|
'
|
2
)],
2
|
'
(|
3
exp[
2
|
'
|
,
1
)
'
(
Then the far field base station antenna gain pattern is given by )
'
,
'
(
)
'
,
'
( 0 I
T
K
I
T h
G
Gxmit
Figure G-2 shows an elevation cut of the pattern while Figure G-3 shows the azimuth cut. Note
the gain is shown in dB in both figures while the angles are in degrees.
Figure G-2: Base Station Transmit Broad beam Elevation Pattern
60 40 20 20 40 60
'
60
50
40
30
20
10
10
G

Figure G-3: Base Station Transmit Broad beam Azimuth Pattern
G.2 Usage of Antenna Gain Pattern Model
The coordinate system described above will be tilted down (with respect to the local horizon) [
degrees ([ o
rotation about x axis) at each base station location in a practical application. Hence
to use the above derived gain pattern in a typical flat earth untilted coordinate system (T, I), a
transformation of variables must be made. The gain in the untilted coordinate system is given by
)]
,
(
'
),
,
(
'
[ I
T
I
I
T
T
xmit
G ; where the elevation and azimuth angle transformation expressions are,
respectively
)
cos(
)
cos(
)
180
sin(
)
sin(
)
180
[cos(
sin
)
,
(
' 1
I
T
[S
T
[S
I
T
T

and
)]
sin(
)
cos(
),
cos(
)
cos(
)
180
cos(
)
sin(
)
180
sin(
[
tan
)
,
(
' 1
I
T
I
T
[S
T
[S
I
T
I

Note that I
T, denote the elevation and azimuth angles respectively in the untilted coordinate
system. For EIRP limit computation the relative pattern is normalized to beam center peak gain
value: Gnorm( ) = Gxmit( )/KG0.
150 100 50 50 100 150
'
10
5
5
10
G

1
APPENDIX H
USE CASE DEVELOPMENT

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 3
LIST OF TABLES.......................................................................................................................... 4
REFERENCES ............................................................................................................................... 5
APPENDIX..................................................................................................................................... 6
Use Case Development........................................................................................... 6

4
LIST OF TABLES

6
APPENDIX
Use Case Development

1
APPENDIX I
FORWARD MODELING RESULTS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 4
LIST OF TABLES.......................................................................................................................... 9
REFERENCES ............................................................................................................................. 10
APPENDIX................................................................................................................................... 11
Forward Modeling Results..................................................................................... 11
I.1 Macro Urban Base Station, GPS C/A-code................................................................... 11
I.1.1 GAV........................................................................................................................ 11
I.1.2 GLN ........................................................................................................................ 16
I.1.3 HPR......................................................................................................................... 21
I.1.4 TIM ......................................................................................................................... 27
I.1.5 CEL......................................................................................................................... 32
I.2 Small Cell Outdoor/Urban Micro Urban Base Station, GPS C/A-code........................ 36
I.2.1 GAV........................................................................................................................ 36
I.2.2 GLN ........................................................................................................................ 41
I.2.3 HPR......................................................................................................................... 46
I.2.4 TIM ......................................................................................................................... 50
I.2.5 CEL......................................................................................................................... 56
I.3 Handsets, GPS C/A-code............................................................................................... 61
I.3.1 GAV........................................................................................................................ 61
I.3.2 GLN ........................................................................................................................ 65
I.3.3 HPR......................................................................................................................... 70
I.3.4 TIM ......................................................................................................................... 73
I.3.5 CEL......................................................................................................................... 77
I.4 Sensitivity Analyses ...................................................................................................... 82
I.4.1 Non-bounding DUTs .............................................................................................. 82
I.4.2 Other GNSS Signal Types ...................................................................................... 87

3
I.4.3 Propagation Models ................................................................................................ 88

4
LIST OF FIGURES
Figure I-1: Maximum Impacted Lateral Distance for Bounding GAV, Macro Urban Base Station
(EIRP = 59 dBm) .......................................................................................................................... 12
Figure I-2: Macro Urban Base Station (EIRP = 59 dBm), Bounding GAV, 1475 MHz.............. 12
Figure I-10: Macro Urban Base Station (EIRP = 59 dBm), Bounding GAV, 1545 MHz............ 15
Figure I-13: Maximum Impacted Lateral Distance for Bounding GLN, Macro Urban Base
Station with EIRP of 59 dBm/sector............................................................................................. 17
Figure I-14: Macro Urban Base Station (EIRP = 59 dBm), Bounding GLN, 1475 MHz ............ 17
Figure I-25: Maximum Impacted Lateral Distance for Bounding HPR, Macro Urban Base Station
with EIRP of 59 dBm/sector......................................................................................................... 22
Figure I-26: Macro Urban Base Station (EIRP = 59 dBm), Bounding HPR, 1475 MHz............. 22

5
Figure I-37: Maximum Impacted Lateral Distance for Bounding TIM, Macro Urban Base
Figure I-38: Macro Urban Base Station (EIRP = 59 dBm), Bounding TIM, 1475 MHz............ 28
Figure I-39: Macro Urban Base Station (EIRP = 59 dBm), Bounding TIM, 1490 MHz............. 28
Figure I-49: Maximum Impacted Lateral Distance for Bounding CEL, Macro Urban Base
Figure I-50: Macro Urban Base Station (EIRP = 59 dBm), Bounding CEL, 1525 MHz............. 34
Figure I-57: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GAV .................... 37
Figure I-58: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GAV, 1475 MHz 37
Figure I-69: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GLN..................... 41
Figure I-70: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GLN, 1475 MHz 42

6
Figure I-81: Small Cell Outdoor/Micro Urban, Bounding HPR................................................... 46
Figure I-82: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding HPR, 1475 MHz. 46
Figure I-93: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding TIM...................... 50
Figure I-94: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding TIM, 1475 MHz . 51
Figure I-100: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding TIM, 1535 MHz 54
Figure I-105: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding CEL ................... 57
Figure I-106: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding CEL, 1525 MHz58

7
Figure I-113: Handset (EIRP = 23 dBm), Bounding GAV .......................................................... 62
Figure I-114: Handset (EIRP = 23 dBm), Bounding GAV, 1620 MHz ....................................... 62
Figure I-121: Handset (EIRP = 23 dBm), Bounding GLN........................................................... 66
Figure I-122: Handset (EIRP = 23 dBm), Bounding GLN, 1620 MHz........................................ 66
Figure I-129: Handset (EIRP = 23 dBm), Bounding HPR ........................................................... 70
Figure I-130: Handset (EIRP = 23 dBm), Bounding HPR, 1620 MHz........................................ 70
Figure I-137: Handset (EIRP = 23 dBm), Bounding TIM............................................................ 73
Figure I-138: Handset (EIRP = 23 dBm), Bounding TIM, 1620 MHz ........................................ 74
Figure I-145: Handset (EIRP = 23 dBm), Bounding CEL ........................................................... 78
Figure I-146: Handset (EIRP = 23 dBm), Bounding CEL, 1620 MHz ........................................ 79

8
Figure I-153: Maximum Impacted Lateral Distance for GAV, Macro Urban Base Station (EIRP
= 59 dBm) ..................................................................................................................................... 83
Figure I-154: Maximum Impacted Lateral Distance for GLN, Macro Urban Base Station (EIRP
= 59 dBm) ..................................................................................................................................... 84
Figure I-155: Maximum Impacted Lateral Distance for HPR, Macro Urban Base Station (EIRP
= 59 dBm) ..................................................................................................................................... 85
Figure I-156: Maximum Impacted Lateral Distance for TIM, Macro Urban Base Station (EIRP =
59 dBm) ........................................................................................................................................ 86
Figure I-157: Maximum Impacted Lateral Distance for CEL, Macro Urban Base Station (EIRP
= 59 dBm) ..................................................................................................................................... 87
Figure I-158: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GLN, 1530 MHz:
(a) GPS C/A-code (b) GLONASS L1C........................................................................................ 88
Figure I-159: Impacted Area for Cellular DUT from 23 dBm EIRP Mobile Device at 1550 MHz
(a) as predicted using free-space propagation model, (b) as predicted using two-ray propagation
model............................................................................................................................................. 89

9
LIST OF TABLES

10
REFERENCES

11
APPENDIX
Forward Modeling Results
This appendix presents forward modeling results. Section I.1 provides results for macro urban
base stations (EIRP = 59 dBm, antenna height = 25 m). Section I.2 provides results for micro
urban/small cell outdoor base stations (EIRP = 40 dBm, antenna height = 6 m). Section I.3
provides results for mobile devices (EIRP = 23 dBm, antenna height = 2 m). These three sections
(Sections I.1, I.2, and I.3) only include results for the most sensitive (bounding) GPS C/A-code
receiver for each frequency in each receiver category (except certified-aviation and spaceborne).
Section I.4 provides the results of a sensitivity analysis. This section explores the variability in
forward modeling results with input assumptions including: (1) percentile of DUTs protected
(e.g., protecting the median DUT vs. the bounding DUT for each frequency and each receiver
category), (2) consideration of other GNSS signal types besides the GPS C/A-code, and (3)
propagation model.
I.1 Macro Urban Base Station, GPS C/A-code
This section presents results for 10-MHz LTE signals broadcast by macro urban base stations
(three sectors with 59 dBm EIRP/sector, 16 dBi antennas at 25 m AGL with 10 deg downtilt),
free-space propagation, and bounding mask.
I.1.1 GAV
Figure I-1 plots the maximum impacted lateral distance for three types of impact to the most
sensitive (bounding) DUT for each frequency: (1) interference results in a 1-dB CNR
degradation (blue), (2) interference results in loss of lock on low-elevation angle satellites (red),
and (3) interference results in loss of lock for high-elevation angle satellites (orange). Figure I-2
to Figure I-12 show the two-dimensional impacted area regions for tested frequencies from 1475
MHz to 1675 MHz.

12
Figure I-1: Maximum Impacted Lateral Distance for Bounding GAV,
Macro Urban Base Station (EIRP = 59 dBm)
Figure I-2: Macro Urban Base Station (EIRP = 59 dBm), Bounding GAV, 1475 MHz
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
1-dB degradation
Loss of Lock, Low Elevation SVs
Loss of Lock, All SVs

13

14

15

16
I.1.2 GLN
MHz to 1675 MHz.

17
Figure I-13: Maximum Impacted Lateral Distance for Bounding GLN,
Macro Urban Base Station with EIRP of 59 dBm/sector
Figure I-14: Macro Urban Base Station (EIRP = 59 dBm), Bounding GLN, 1475 MHz
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
1-dB degradation

18

19

20

21
I.1.3 HPR
MHz to 1675 MHz.

22
Figure I-25: Maximum Impacted Lateral Distance for Bounding HPR, Macro Urban Base
Station with EIRP of 59 dBm/sector
Figure I-26: Macro Urban Base Station (EIRP = 59 dBm), Bounding HPR, 1475 MHz
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
2
10
3
10
4
10
5
10
6
Maximum
Impacted
Lateral
Distance
1-dB degradation

23

24

25

26

27
I.1.4 TIM
MHz to 1675 MHz.
Figure I-37: Maximum Impacted Lateral Distance for Bounding TIM,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
1-dB degradation

28
Figure I-38: Macro Urban Base Station (EIRP = 59 dBm), Bounding TIM, 1475 MHz

29

30

31

32
I.1.5 CEL
degradation (blue), (2) interference results in loss of lock on -10 dB satellites1
(red), and (3)
interference results in loss of lock for high-elevation angle satellites (orange). Figure I-50 to
Figure I-56 show the two-dimensional impacted area regions for tested frequencies from 1525
MHz to 1675 MHz. (Note that for the other LTE frequencies tested, there was no impact).
1
All of the plots in this Appendix that use the loss-of-lock interference level towards nominally powered satellites
are referred to as “all SVs” and towards -10 dB satellites as “low elevation SVs” since a typical DUT antenna
exhibited 10 dB less gain towards low elevation angles as compared to its gain at zenith. However, these curves
should be interpreted differently for CEL devices since as described in the main body of this report, CEL antennas
were modeled as isotropic. The “all SVs” curve can still be interpreted as the interference level that would result in
loss of tracking of all satellites, but the “low elevation SVs” curve should be interpreted as the interference level that
would result in the device losing lock on satellite signals that are attenuated by 10 dB due, e.g., to line-of-sight
blockage.

33
Figure I-49: Maximum Impacted Lateral Distance for Bounding CEL,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
Maximum
Impacted
Lateral
Distance
1-dB degradation

34
Figure I-50: Macro Urban Base Station (EIRP = 59 dBm), Bounding CEL, 1525 MHz

35

36
I.2 Small Cell Outdoor/Urban Micro Urban Base Station, GPS C/A-code
This section presents results for 10-MHz LTE signals broadcast by small cell outdoor/micro
urban base stations (one sector with 40 dBm EIRP, 5 dBi antenna at 6 m AGL), free-space
propagation, bounding mask.
I.2.1 GAV

37
Figure I-57: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GAV
Figure I-58: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm),
Bounding GAV, 1475 MHz
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
0
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
1-dB degradation

38

39

40

41
I.2.2 GLN
Figure I-69: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding GLN
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
0
10
1
10
2
10
3
10
4
Maximum
Impacted
Lateral
Distance
1-dB degradation

42
Bounding GLN, 1475 MHz

43

44

45

46
I.2.3 HPR
Figure I-81: Small Cell Outdoor/Micro Urban, Bounding HPR
Bounding HPR, 1475 MHz
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
1-dB degradation

47

48

49

50
I.2.4 TIM
Figure I-93: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding TIM
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
0
10
1
10
2
10
3
10
4
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation

51
Bounding TIM, 1475 MHz

52

53

54

55

56
I.2.5 CEL

57
Figure I-105: Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding CEL
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
0
10
1
10
2
10
3
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation

58
Bounding CEL, 1525 MHz

59

60

61
Figure X. Small Cell Outdoor/Micro Urban (EIRP = 40 dBm), Bounding CEL
I.3 Handsets, GPS C/A-code
This section presents results for 10-MHz LTE signals broadcast by handsets (EIRP = 23 dBm,
isotropic antenna at 2 m AGL), free-space propagation, bounding mask.
I.3.1 GAV

62
Figure I-113: Handset (EIRP = 23 dBm), Bounding GAV
Figure I-114: Handset (EIRP = 23 dBm), Bounding GAV, 1620 MHz
1620 1625 1630 1635 1640 1645 1650 1655 1660
Frequency (MHz)
10
0
10
1
10
2
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation

63

64

65
I.3.2 GLN

66
Figure I-121: Handset (EIRP = 23 dBm), Bounding GLN
Figure I-122: Handset (EIRP = 23 dBm), Bounding GLN, 1620 MHz
1620 1625 1630 1635 1640 1645 1650 1655 1660
Frequency (MHz)
10
-1
10
0
10
1
10
2
10
3
Maximum
Impacted
Lateral
Distance
1-dB degradation

67

68

69

70
I.3.3 HPR
Figure I-129: Handset (EIRP = 23 dBm), Bounding HPR
Figure I-130: Handset (EIRP = 23 dBm), Bounding HPR, 1620 MHz
1620 1625 1630 1635 1640 1645 1650 1655 1660
Frequency (MHz)
10
0
10
1
10
2
10
3
10
4
Maximum
Impacted
Lateral
Distance
1-dB degradation

71

72

73
I.3.4 TIM
Figure I-137: Handset (EIRP = 23 dBm), Bounding TIM
1620 1625 1630 1635 1640 1645 1650 1655 1660
Frequency (MHz)
10
0
10
1
10
2
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation

74
Figure I-138: Handset (EIRP = 23 dBm), Bounding TIM, 1620 MHz

75

76

77
I.3.5 CEL

78
Figure I-145: Handset (EIRP = 23 dBm), Bounding CEL
1620 1625 1630 1635 1640 1645 1650 1655 1660
Frequency (MHz)
10
-1
10
0
10
1
10
2
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation

79
Figure I-146: Handset (EIRP = 23 dBm), Bounding CEL, 1620 MHz

80

81

82
I.4 Sensitivity Analyses
Sections I.1 to I.3 presented results that were applicable for:
• The most sensitive (bounding) DUT of each receiver category for each frequency.
• GPS C/A-code tracking.
• Environments where the free-space path loss model is appropriate.
This section examines sensitivity of the results to these assumptions.
I.4.1 Non-bounding DUTs
Figure I-153 to Figure I-157 provide results for the maximum impacted lateral distance for both
the bounding and median DUTs for each receiver category and frequency. The results in these
plots are applicable for a single macro urban base station (EIRP = 59 dBm) and for reception of
the GPS C/A-code. It is important to note that “median” pertains only to the set of devices tested

83
at WSMR for each category, and these results should not be interpreted as being applicable to the
“median” of fielded GPS/GNSS receivers. As expected, the impacted distances are significantly
smaller for the median vs bounding DUTs. Loss-of-lock curves are only included in the HPR
plot (Figure I-155) since the median DUTs for all of the other receiver categories did not lose
lock on either the nominal or -10 dB satellite signals at any tested frequency.
Figure I-153: Maximum Impacted Lateral Distance for GAV,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
(m)
1-dB degradation (bounding)
Loss of Lock, Low Elevation SVs (bounding)
Loss of Lock, All SVs (bounding)
1-dB degradation (median)

84
Figure I-154: Maximum Impacted Lateral Distance for GLN,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
(m)

85
Figure I-155: Maximum Impacted Lateral Distance for HPR,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
10
6
Maximum
Impacted
Lateral
Distance
(m)
Loss of Lock, Low Elevation SVs (median)
Loss of Lock, All SVs (median)

86
Figure I-156: Maximum Impacted Lateral Distance for TIM,
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
10
5
Maximum
Impacted
Lateral
Distance
(m)

87
Figure I-157: Maximum Impacted Lateral Distance for CEL,
I.4.2 Other GNSS Signal Types
Sections I.1 through I.3 presented results only for GPS C/A-code reception. Oftentimes, DUTs
that were capable of tracking other GNSS signal types were more sensitive to interference for the
processing of the other signals. As an example, Figure I-158 shows the impacted areas for a
small cell outdoor base station operating at 1530 MHz for the most-sensitive GLN DUT. The top
figure (a) is for GPS C/A-code, and the bottom figure (b) for GLONASS L1C.
1450 1500 1550 1600 1650 1700
Frequency (MHz)
10
1
10
2
10
3
10
4
Maximum
Impacted
Lateral
Distance
(m)

88
Bounding GLN, 1530 MHz: (a) GPS C/A-code (b) GLONASS L1C
I.4.3 Propagation Models
Sections I.1 through I.3 presented forward modeling results that presumed free-space
propagation. Depending on the operational environment, these results can be excessively
optimistic or pessimistic.
When there is a clear line-of-sight path between the transmitting and receiving antennas and
additionally a strong ground reflection, the free-space propagation model can yield optimistic
results (i.e., the maximum impacted distance may be greater than predicted). In such
environments, a two-ray path model may provide greater accuracy in predicting received power
levels. Figure I-159 provides an example of the difference in impacted areas between free-space
propagation and two-ray propagation. The two-ray result presumes that the reflecting ground
surface is smooth concrete. As a second example, received power levels from an LTE base
station in a rural area near Las Vegas were observed at power levels significantly greater than
(a)
ITM =-63.5
LOL
L
=-43.1
LOL
H
=-33.1
0 200 400 600 800 1000 1200 1400
Distance from Base Station (m)
0
100
200
Height
(m)
(b)

89
predicted using the free-space path loss model at lateral distances of up to 9 km.2
The base
station antenna height was 18 m and the received power was measured at a height of around 2 m.
Figure I-159: Impacted Area for Cellular DUT from 23 dBm EIRP
Mobile Device at 1550 MHz (a) as predicted using free-space propagation model,
(b) as predicted using two-ray propagation model
When there is significant blockage between the transmitting and receiving antennas, the free-
space propagation model can at times yield pessimistic results (i.e., the maximum impacted
distance may be far smaller than predicted). For example, received power measurements from an
LTE base station in a dense urban area within Las Vegas were typically observed to be 10 – 20
dB below those levels predicted using free-space propagation for lateral distances of up to 2 km.3
The base station antenna height was 72 m and the received power was measured at a height of
around 2 m. It should be noted that, even though the received power was most frequently less
than predicted using free-space, in some locations far from the base station ( 1 km) power
levels were observed to be as high as or exceeding the free-space propagation model predictions.
Such observations were much more frequent within 500 km of the base station.
Given the prevalence of line-of-sight blockages in urban and dense suburban areas, the free-
space results in Sections I.1 through I.3 are likely to be significantly pessimistic for distances
2
See p. 110 of https://ecfsapi.fcc.gov/file/7021690471.pdf.
3
See p. 107 of https://ecfsapi.fcc.gov/file/7021690471.pdf.
ITM =-46.1
LOL
L
= N/A
LOL
H
= N/A
0 10 20 30 40
Distance from Mobile Device (m)
0
10
20
30
40
Height
(m)
(a)
ITM =-46.1
LOL
L
= N/A
LOL
H
= N/A
0 10 20 30 40 50 60 70 80
Distance from Mobile Device (m)
0
10
20
30
40
Height
(m)
(b)

90
exceeding a few hundred meters for all azimuths and significantly pessimistic for some azimuths
at shorter distances. Importantly, however, LTE macro base stations may be separated by
distances as small as 250 m and small cell outdoor base stations even shorter distances (see Table
3-13). Over several hundred meters in all environments including dense urban, the free-space
propagation and two-ray propagation models are appropriately conservative models to protect
GPS/GNSS receivers from interference from adjacent band mobile broadband systems.

1
APPENDIX J
INVERSE MODELING RESULTS

2
TABLE OF CONTENTS
LIST OF FIGURES ........................................................................................................................ 4
LIST OF TABLES.......................................................................................................................... 6
REFERENCES ............................................................................................................................... 7
APPENDIX..................................................................................................................................... 8
Bounding and Median EIRP Tolerance Masks (ETM) for Single Transmitter....... 8
J.1 GPS L1 C/A ETMs ......................................................................................................... 8
J.1.1 Micro Urban Deployment......................................................................................... 8
J.1.1.1 Bounding ETMs............................................................................................. 8
J.1.1.1.1 FSPL Propagation Model............................................................................... 8
J.1.1.1.2 2-Ray Path loss model ................................................................................. 10
J.1.1.2 Median ETMs .............................................................................................. 11
J.1.1.2.1 FSPL Propagation Model............................................................................. 12
J.1.2 Macro Urban Deployment...................................................................................... 14
J.1.2.1 Bounding ETMs........................................................................................... 15
J.1.2.2 Median ETMs .............................................................................................. 18
J.2 All GNSS services......................................................................................................... 22
J.2.1 Micro Urban Deployment....................................................................................... 22
J.2.1.1 Bounding ETMs........................................................................................... 22
J.2.1.2 Median ETMs .............................................................................................. 26

3
...................................................................................... 29
J.2.2.1 Bounding ETMs........................................................................................... 30
J.2.2.2 Median ETMs .............................................................................................. 33

4
LIST OF FIGURES
Figure J-1: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 500 m............... 8
Figure J-2: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 100 m............... 9
Figure J-3: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 10 m................. 9
Figure J-4: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 500 m ............ 10
Figure J-5: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 100 m ............ 10
Figure J-6: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 10 m .............. 11
Figure J-7: GPS L1 C/A, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 500 m ................ 12
Figure J-8: GPS L1 C/A, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 100 m ................ 12
Figure J-9: GPS L1 C/A, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 10 m .................. 13
Figure J-10: GPS L1 C/A, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 500 m.............. 13
Figure J-11: GPS L1 C/A, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 100 m.............. 14
Figure J-12: GPS L1 C/A, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 10 m................ 14
Figure J-13: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 1000 m........ 15
Figure J-14: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 100 m.......... 16
Figure J-15: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 10 m............ 16
Figure J-16: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 1000 m ....... 17
Figure J-17: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 100 m ......... 17
Figure J-18: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 10 m ........... 18
Figure J-19: GPS L1 C/A, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 1000 m............ 19
Figure J-20: GPS L1 C/A, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 100 m.............. 19
Figure J-21: GPS L1 C/A, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 10 m................ 20
Figure J-22: GPS L1 C/A, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 1000 m........... 21
Figure J-23: GPS L1 C/A, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 100 m............. 21
Figure J-24: GPS L1 C/A, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 10 m............... 22
Figure J-25: All GNSS, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 500 m............... 23
Figure J-26: All GNSS, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 100 m............... 23
Figure J-27: All GNSS, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 10 m................. 24
Figure J-28: All GNSS, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 500 m .............. 25
Figure J-29: All GNSS, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 100 m .............. 25
Figure J-30: All GNSS, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 10 m ................ 26
Figure J-31: All GNSS, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 500 m .................. 27
Figure J-32: All GNSS, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 100 m .................. 27
Figure J-33: All GNSS, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 10 m .................... 28
Figure J-34: All GNSS, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 500 m.................. 28
Figure J-35: All GNSS, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 100 m.................. 29
Figure J-36: All GNSS, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 10 m.................... 29

5
Figure J-37: All GNSS, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 1000 m ............ 30
Figure J-38: All GNSS, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 100 m .............. 31
Figure J-39: All GNSS, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 10 m ................ 31
Figure J-40: All GNSS, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 1000 m............ 32
Figure J-41: All GNSS, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 100 m.............. 32
Figure J-42: All GNSS, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 10 m................ 33
Figure J-43: All GNSS, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 1000 m................ 34
Figure J-44: All GNSS, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 100 m.................. 34
Figure J-45: All GNSS, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 10 m.................... 35
Figure J-46: All GNSS, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 1000 m............... 36
Figure J-47: All GNSS, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 100 m................. 36
Figure J-48: All GNSS, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 10 m................... 37

6
LIST OF TABLES

8
APPENDIX
Bounding and Median EIRP Tolerance Masks (ETM) for Single Transmitter
J.1 GPS L1 C/A ETMs
J.1.1 Micro Urban Deployment
J.1.1.1 Bounding ETMs
J.1.1.1.1 FSPL Propagation Model
Figure J-1: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 500 m

9

10
J.1.1.1.2 2-Ray Path loss model
Figure J-4: GPS L1 C/A, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 500 m

11
J.1.1.2 Median ETMs

12
Figure J-7: GPS L1 C/A, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 500 m

13
Figure J-10: GPS L1 C/A, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 500 m

14
J.1.2 Macro Urban Deployment

15
Figure J-13: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 1000 m

16

17
Figure J-16: GPS L1 C/A, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 1000 m

18
J.1.2.2 Median ETMs

19
Figure J-19: GPS L1 C/A, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 1000 m

20

21
Figure J-22: GPS L1 C/A, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 1000 m

22
J.2 All GNSS services
J.2.1 Micro Urban Deployment

23
Figure J-25: All GNSS, Micro Urban, Bounding EIRP Mask: FSPL, dStandoff = 500 m

24

25
Figure J-28: All GNSS, Micro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 500 m

26
J.2.1.2 Median ETMs

27
Figure J-31: All GNSS, Micro Urban, Median EIRP Mask: FSPL, dStandoff = 500 m

28
Figure J-34: All GNSS, Micro Urban, Median EIRP Mask: 2-Ray, dStandoff = 500 m

29

30
Figure J-37: All GNSS, Macro Urban, Bounding EIRP Mask: FSPL, dStandoff = 1000 m

31

32
Figure J-40: All GNSS, Macro Urban, Bounding EIRP Mask: 2-Ray, dStandoff = 1000 m

33
J.2.2.2 Median ETMs

34
Figure J-43: All GNSS, Macro Urban, Median EIRP Mask: FSPL, dStandoff = 1000 m

35

36
Figure J-46: All GNSS, Macro Urban, Median EIRP Mask: 2-Ray, dStandoff = 1000 m

37

1
APPENDIX K
SPACEBORNE AND SCIENCE
GNSS APPLICATIONS

2
SPACEBORNE AND SCIENCE-APPLICATIONS
Overview
This section of the report describes the analysis and evaluation of a proposed LTE base station
network’s interference to space-based receivers. A comprehensive assessment on GNSS
receivers, used in various applications, supporting NASA’s portfolio of missions is also
addressed. However, the emphasis of this section is on the assessment to GNSS receivers used
for science applications.
The following evaluation assesses the impact to space-based GNSS receivers performing radio
occultation (“RO”) measurements (a scientific application of GNSS) of the ionosphere,
stratosphere, and the troposphere. RO measurements, coupled with traditional methodologies for
weather prediction, provide weather and science data observations from ocean areas, the
atmosphere, and other natural phenomena, which have improved accuracy and predictability of
weather forecasts by as much as two days.
Specifically, NASA’s assessment focuses on the most recent developed RO receiver, called the
TriG, developed by the NASA/Jet Propulsion Laboratory (JPL). The TriG is the newest RO
receiver of the BlackJack class of GNSS receivers. The increase in performance by these
receivers is partially due to the TriG’s ability to receive all GNSS signals: GPS, Galileo,
GLONASS, Compass, as well as other future navigation signals (QZSS, DORIS, etc.).
Radio Frequency Interference (RFI) is an utmost problem when GNSS signals are being used for
science applications. When RFI occurs at low ray heights, the GNSS signal is defocused by tens
of dB, and the signal-to-noise ratio (SNR) is already being measured in a marginal zone. In fact,
in this already marginal zone, tracking loops cannot be closed and the captured data is running
open loop. The spatially correlated noise can bias the captured data and affect the climate
record, in addition to reducing weather forecast accuracy over the USA where the LTE base
stations would be deployed.
This assessment demonstrates the effect of RFI generated by the ground-based LTE network.
Several iterations of the modeling and simulation (MS) runs were performed to more
accurately model the presumed network deployment of the interfering network. The MS
scenarios estimate the receive interference levels to the TriG, utilizing specific mission
parameters, and comparing them against interference limits/thresholds obtained through anechoic
chamber testing described under Section 3.
Background
Radio Occultation (GNSS-RO)
GNSS-RO is the measurement of GNSS signals as they are refracted by the atmosphere. RO is a
relatively new method for the indirect measurement of temperature, pressure and water vapor in
the stratosphere and the troposphere, and of charged particles in the ionosphere. These
measurements are made from specifically designed GNSS receivers on-board a Low-Earth-Orbit

3
(LEO) satellite. The techniques utilize the unique radio signals continuously transmitted by the
GNSS satellites (GPS, GLONASS, Galileo, etc.) orbiting the Earth at an approximate altitude of
20,000 km above the surface. The GNSS radio signals are influenced both by the electron
density in the ionosphere and by the variations of temperature, pressure and water vapor in the
atmosphere which are used in meteorology and climate science. RO measurements are also used
to derive various ionospheric parameters (Total Electron Content (TEC), Electron Density
Profiles (EDP), L-band scintillation, etc.) for understanding earth and space weather dynamics.
Figure K-1: Progression of Tangent Point for a Setting (Descending) Occultation
From the point of view of a LEO satellite (at an altitude of 700-800 km), the GNSS satellites
continually rise above, or set behind, the horizon of the Earth. During these so-called radio
occultations, where the GNSS and the LEO satellite are just able to see each other through the
atmosphere, the GNSS signals will be slightly delayed and their ray path slightly bent (refracted)
on the way through the layers of the atmosphere (see Figure K-1). The excess range increases
as the ray propagates through denser mediums and water vapor at lower altitudes. This delay is a
function of density (n/V), which is related to temperature by the ideal gas law (Equation K-1),
P*V = n*R*T
Equation K-1: Equation Used to Translate Refractivity to Atmospheric Temperature
A typical occultation sounding will last one (1) to two (2) minutes, and during this time the LEO
satellite will receive signals where the ray paths have different minimum distances to the surface
of the Earth, from zero up to approximately 100 km. The GNSS satellites transmit on multiple
frequencies, and with a receiver rate of 50 Hz this will yield around 6000 rays, making up a
profile of excess phases (actual path minus straight-line path) through the lowest 100 km of the
atmosphere.
 UCAR

4
The residual positioning error and determination of time delays (see Figure K-2), derived from
the measurements taken during a RO event, are key parameters in the obtaining the temperature,
pressure, and water vapor characteristics of the atmosphere at different heights. Given sub-mm
measurement precision, RO can determine atmospheric temperature profiles to 0.1 – 0.5 Kelvin
(K) accuracy from 8 - 25 km height levels. These are unprecedented levels of accuracy for global
measurements.
Figure K-2: Straight Line versus Actual Path of GNSS Signal
NASA has several radio occultation receivers in its portfolio, including the Integrated GPS
Occultation Receiver (IGOR), the IGOR+, and a more recently developed receiver called the
TriG receiver.
NASA/JPL TriG Receiver Overview
The NASA/JPL developed TriG receiver functions as a multi-function GNSS receiver. This
single receiver has multiple antenna inputs and can be configured to operate in a navigation
capacity, as well as, simultaneously, in a scientific measurement role.
In its traditional function, coupled with choke ring antenna (see
Figure K-3), the TriG serves as a device for space vehicle navigation
and precise orbit determination (POD). The receiver provides
accurate information to space vehicle operators on position, velocity,
and time.
Configured in a scientific measurement mode, the TriG,
coupled with a series of specially designed antenna arrays,
performs RO measurements of GNSS signals. TriG receiver hardware is able to receive all
GNSS signals: GPS, Galileo, GLONASS, Compass, as well as other navigation signals (QZSS,
DORIS, etc.). This capability increases the number of RO measurements that can be made
during any given orbit.
 UCAR
Figure K-3: Typical Choke
Ring Antenna

5
Additional information on TriG can be found in a document titled, “TriG - A GNSS Precise
Orbit and Radio Occultation Space Receiver”, written by the Jet Propulsion Laboratory and
California Institute of Technology1
.
TriG Pre-Select Filter
Much akin to high-precision (HP) GPS receivers, the TriG has been designed with a wide front-
end receiver filter. This wider pre-select filter can be derived from the ITMs developed by the
DOT (see Section 3). Although the DOT developed ITMs for each of the six (6) categories of
GPS receivers using bounding results, NASA specifically tested two (2) spaced-based receivers
during the anechoic chamber tests.
The pre-select filter contains two 150 MHz-wide pass bands. The first covers the L5 and L1
bands. The second pass band covers only 1 but was designed with the same bandwidth to
equalize filter delay changes with temperature. This is important for the ionospheric
measurements, a prime product of the COSMIC-2BB mission. The TriG also has 2nd
stage
narrow band filters that are centered around the GPS L1 and L2 bands.
Receivers are purposely designed to have a wider bandwidth for both HP and the TriG receivers.
The wider bandwidth front-end filter takes advantage of:
• The ability to track all current and future GNSS L-band signals:
- GPS
- Galileo (Europe)
- GLONASS (Russia)
- Compass/BeiDou (China)
- QZSS (Japan)
- NaVIC (formerly, IRNSS) (India)
- DORIS (France)
- GPS augmentation systems operating on mobile satellite service (MSS) frequency
allocations, and
- Other future GNSS constellations.
• Avoiding the disadvantages that narrow filters with sharp cutoffs produce, such as:
- Distorted ranging code transitions
- Introduction of inter-signal biases which vary with temperature and Doppler
- Increased insertion loss that degrade SNR, and
- Phase and delay distortion across signal band.
1
http://authors.library.caltech.edu/21729/1/Esterhuizen2009p12347Proceedings_Of_The_22Nd_International_Techn
ical_Meeting_Of_The_Satellite_Division_Of_The_Institute_Of_Navigation_(Ion_Gnss_2009).pdf

6
In addition to the typical advantages afforded to HP receivers that are designed with wider front-
end bandwidth filters, the TriG gains additional benefits for employing wide bandwidth filters
by:
• Avoiding extensive development cost and time,
• Avoiding the additional cost for pre-flight testing,
• Avoiding the additional costs associated with size and mass restrictions of flight
instrument, and
• Leveraging on advanced techniques such as:
- Oversampling the GNSS signal and use of narrow-lag correlators for better precision,
and
- On-receiver multipath mitigation techniques.
Upcoming TriG Missions
TriG receivers will be flown on the next generation radio occultation capable satellites as part of
the COSMIC-2B2
mission, which is sponsored by several U.S. federal agencies and NASA
international partners. The COSMIC-2 mission is broken down into two (2) sub-missions, which
will deploy six (6) satellites each. Table K-1 displays the upcoming missions where the TriG
receiver will be deployed.
Note: The list of missions in Table K-1 depicts the known missions, as of the writing of this
report. As NASA continues to develop partnerships with other International Space Agencies and
other U.S. Federal partners, coupled with the success of integrating RO measurements into the
weather prediction models, it should be noted that this list may change in the future.
Mission Launch Date TriG Function
Deep Space Atomic Clock (DSAC) Jul-18
Precise clock validation
Timing
POD
Constellation Observing System
for Meteorology, Ionosphere and
Climate (COSMIC)-2 (A) - 6
satellites
Jul-18
RO
SWO
Gravity Recovery and Climate
Experiment (GRACE) Follow-On
Feb-18
Micron ranging
POD
RO
COSMIC-2B (B) - 6 satellites RO
2
Reference, http://www.cosmic.ucar.edu/cosmic2/

7
2020 (pending
funding) *
SWO
Sentinel-6A and B (2 launches)
2020 and
2025*
RO
POD
Surface Water and Ocean
Topography (SWOT)
2021* POD
NASA-ISRO Synthetic Aperture
Radar (NISAR)
2021* POD
Table K‑1: TriG Mission List (as of Oct-2017)
Table Legend:
POD – Precision Orbit Determination
RO – Radio Occultation
SWO – Space Weather Observation
* Tentative mission launch year
Other Scientific Applications of GNSS
GNSS technology has become an essential tool to monitor and improve our understanding of
earth systems, including weather monitoring and solid earth hazards such as earthquakes and
volcanic activity. This knowledge of our environment and its changes is also used for resource
management and protection, and environmental impact mitigation. Some examples of the use of
GNSS to improve our knowledge of the Earth are determining the atmosphere’s water content,
improving the accuracy of weather forecasts, enabling ocean topography measurements to
determine currents and secular changes in sea height. Ground based GNSS networks are also
playing an increasingly prominent role to monitor ground movement to identify potential
conditions that may precede Earthquakes and volcanic activity. In addition, some insurance
companies use GNSS-based maps of accumulated tectonic strain to predict risk. The same data
are used by other government agencies beyond NASA. GNSS technology assists NASA
scientists in understanding the physical characteristics of the earth and its atmosphere, and
changes over time. NASA scientists use GPS science receivers, in combination with other
measurement techniques such as laser ranging and radar altimeters, to monitor the changes in
Earth’s surface, sea level height, and atmospheric measurements and provide precise knowledge
of Earth’s shape and rotation.
As the scientific community continues to embrace leveraging on GNSS, additional techniques
have been developed to measure and monitor earth and space weather phenomena. These
techniques take advantage of:
• Existing development and deployment of satellite constellations, thereby, saving money
in developing and deploying a separate constellation for science signals;

8
• Existing satellite constellations providing signals known and consistent position
determinations all around the Earth; and
• GNSS signals transmit precise time and positioning information continuously in all
weather conditions.
Ground-based GNSS Receivers Used for Integrated Precipitable Water Measurements
This recently developed technique in performing atmospheric observations utilizes ground-based
GNSS receivers that employ zenith (away from earth) pointing antennas to measure GNSS
signals. As the GNSS satellite comes into view of the antenna overhead, the amount of
measured delay of the signal due to water vapor in the atmosphere can be measured and
attributed to specific weather conditions. As a meteorological application, ground-based GNSS
receiver data is used to derive the Integrated Precipitable Water which is fed into the Numerical
Weather Prediction model. This data is complementary to the space-based data (RO), and
together, they provide valuable ionospheric information for space weather specification and
forecasting.
In this system, commercially available HP GPS/GNSS receivers are typically utilized and the
data is fed into post-processing algorithms to determine the precipitable water vapor content of
the atmosphere.
Although NASA utilizes such systems to correlate the water vapor data with RO measurements
to more accurately predict weather phenomena, NASA did not perform any specific assessments
to these systems under the DOT ABC Assessment. Since NASA leverages on commercial HP
GPS/GNSS receivers to perform these measurements, any such protection criteria and separation
distances afforded to the HP category of receivers under Section 3 will be applicable to locations
where ground-based GPS/GNSS receivers are used for metrology.
Reflectometry (GNSS-R)
In addition to radio occultation and ground-based GNSS measurements, measuring the
characteristics of Earth and bodies of water through a technique called “reflectometry” (GNSS-
R) is also valuable application for science and weather.
For example, NASA’s Cyclone Global Navigation Satellite System (CYGNSS) mission,
consisting of eight (8) small satellite observatories, which was launched in 2016, will make
frequent and accurate measurements of ocean surface winds throughout the life cycle of tropical
storms and hurricanes. In addition to using GNSS signals for satellite navigation, each satellite
observatory can measure four (4) separate GNSS signals at the specular reflection points on the
ocean to obtain information about ocean surface roughness. Ocean surface roughness is
correlated to surface wind speed. The CYGNSS data will enable scientists to probe key air-sea
interaction processes that take place near the core of storms, which are rapidly changing and play
a critical role in the genesis and intensification of hurricanes.
Spacecraft equipped with GNSS-R systems receive a direct GNSS signal, as well as a “reflected”
GNSS signal from the Earth’s surface. The direct signal is transmitted from a GNSS satellite
and received by a zenith pointing antenna onboard the spacecraft, while the reflected signal is
received by the two (2) nadir (towards the earth) pointing antennas. If the surface is perfectly

9
smooth, the specular reflection point is the location on the surface where all of the scattering
originates. In comparison, if the surface is roughened (e.g., due to over the surface wind speed),
the scattering of the GNSS signal originates from a diffuse region called the glistening zone
around the specular point. Figure K-4 demonstrates a pictorial of GNSS-R concept of
operations.
Figure K-4: GNSS-R Concept of Operations (Image Credit – University of Michigan)
Algorithms measure the amplitude of the reflected signal versus
delay and Doppler shift. If the surface is smooth, nearly all
power originates at the specular reflection point. If the surface is
rough, there are reflections from facets separated from the
specular points. Those reflections have more delay, and a spread
of Doppler shifts. An example of Delay Doppler Maps for 2, 7,
and 10 meter per second (m/s) wind speeds [top to bottom] is
shown in Figure K-5.
[Illustration Note: The images show how progressively stronger
wind speeds, and therefore progressively rougher sea surfaces,
produce a weaker maximum signal (at the top of the “arch”) and
a scattered signal along the arch that is closer in strength to the
maximum. A perfectly smooth surface would produce a single
red spot at the top of the arch.3
Image credit: University of
Michigan.]
In addition to weather forecasting (e.g., cyclonic and hurricane
activity), GNSS-R has shown promises to predict other Earth surface phenomena relating to
3
Additional information on CYGNSS can be found at the following weblink to the NASA CYGNSS Mission site:
https://www.nasa.gov/cygnss/overview.
Figure K-5: Example
Delay Doppler Maps

10
bodies of water. NASA scientists are exploring the capability of GNSS-R receivers to monitor
and anticipate:
• Coastal tidal surges,
• River and lake overflows,
• Flood plains,
• Water surges beneath foliage canopies (e.g., swamps and mangroves),
• Potential dyke, reservoir, and dam exceedances, and
• Many more areas that may be impacted due to watershed anomalies.
Since GNSS-R is a relatively new technique used as a scientific application of GNSS, NASA
was unable to obtain a GNSS-R receiver to be tested during the testing phases (anechoic
chamber or conducted) of the DOT ABC Assessment. Therefore, the effects of adjacent band
LTE operations to GNSS-R are currently unknown.
Geodesy/Geodetics
Geodesy or geodetics is the science of accurately measuring and obtaining data to understand the
properties of the Earth. In this scientific discipline, observations are performed to obtain
information on the Earth’s geometric shape, orientation (relative to Earth’s axis and the sun),
crustal motion, oceanic tides, and Earth’s gravitational field. Since these Earth properties are
continuously changing, measurements are taken with respect to time. To ensure stability and
consistency in these measurements, scientists leverage on a known and constant signal source,
like GPS and other GNSS signals, where accurate three-dimensional positioning attributes and
timing can be obtained.
In order to accurately measure these Earth properties, commercially available HP GPS/GNSS
receivers are typically utilized at fixed locations on the Earth’s surface. Scientifically measured
data is fed into post-processing algorithms to determine the three-dimensional positioning (in
some cases such as for earthquake monitoring, accuracy levels must be down to millimeters).
The variations of the receiver positons with time are compared with physical models to
determine inter-earthquake strain accumulation, earth orientation parameters, etc.
Since NASA leverages on commercial HP GPS/GNSS receivers to perform these scientific
measurements, any such protection criteria and separation distances afforded to the HP category
of receivers under Section 3 will be applicable to locations where ground-based GPS/GNSS
receivers are used for geodesy/geodetic science.
Real-time Response to Natural Hazards
The U.S. Geological Survey (USGS) mandate includes monitoring and responding to natural
hazards due to earthquake faults, volcanoes, landslides, and tsunamis. Many of these
applications depend critically upon GNSS data and require the broadest available spectrum of
GNSS signals, including side bands, to achieve the highest station position accuracy and
precision possible in real-time. In particular, the USGS Earthquake Hazards Program, in
collaboration with universities and other governmental and private agencies, is developing an
earthquake early warning (EEW) system, called ShakeAlert (USGS OFR # 2014-1097).

11
ShakeAlert will issue life-critical alerts of impending strong ground motion for public safety and
emergency response in the event of a major earthquake.
Seismometers often go off-scale when recording seismic waves during large and great
earthquakes, and magnitudes calculated from the earliest portion of a seismogram can be
significantly underestimated. In an EEW system this leads, in turn, to underpredicted ground
shaking. GNSS data have the unique capability to measure large displacements reliably at the
centimeter level without going off-scale, thus augmenting seismic data to enable accurate
magnitude estimates for M7+ earthquakes.
In order to do this, the GNSS component of ShakeAlert requires real-time, uninterrupted GNSS
signals without interference at all times from a broadly distributed network with stations near the
earthquake faults. Even brief outages due to RFI can significantly affect the precision of the
GNSS observations and degrade the performance of the system. In addition to data from GNSS
stations in the earthquake source region themselves, calculating their absolute positions in real-
time also requires the continuous availability of data from GNSS stations outside the affected
region in order to generate real-time clock correction streams. It can take up to 15 minutes for a
receiver to recover from its own loss of lock and/or loss of the correction streams and resume
production of the position streams. During this recovery time, a receiver cannot contribute to
EEW because it will not output reliable positions, which are needed for rapid earthquake and
ground motion characterization. Since early warnings must be sent within seconds of the onset
of an earthquake to be useful, such delays would compromise the EEW system. The growing
USGS volcano alert system could be similarly affected by RFI-related outages.
In addition to the use of GNSS-derived positions for earthquake response, the USGS seismic
networks rely on the GNSS constellation L1 signal in order to mitigate seismometer clock drift.
To implement this, seismic instruments typical of those used by USGS networks include a GPS
chipset centered at 1550 MHz. The accuracy of time tags assigned to seismic data recorded in
the field is critical for determining accurate earthquake locations, but GPS loss of lock leads to
clock drift. This is particularly detrimental for EEW, in which the correct location of an
earthquake must be determined within seconds in order to maximize the warning time that can be
given to affected populations.
Other NASA Applications of GNSS Receivers
Statistically, nearly 60% of projected worldwide space missions present-2027 will operate in
LEO. Additionally, 35% of space missions that will operate at higher altitudes will remain at or
below Geostationary-Earth-Orbit (GEO). Therefore, approximately 95% of projected worldwide
space missions over the next 20 years will operate within the GNSS service envelope and will
rely on GNSS for space activities associated with navigation, POD, science, and other
applications.
The following sections describe the uses of GNSS receivers that support various NASA
missions.

12
Note: Although the following applications, coupled with the science applications of GNSS (in
above sections) provide for a comprehensive list of NASAs’ uses
of GNSS, it should be noted that this does not provide a full
complement of NASA’s uses of GNSS receivers. Other uses for
day-to-day operations, NASA security, fire and rescue, etc.,
typically utilize GLN receivers, which are addressed in Section 3.
Therefore, any constraints to LTE operations required to protect
GLN devices will be applicable to these NASA functions.
Aviation Systems
NASA’s Aeronautical Research Mission Directorate operates NASA owned, maintained, and
operated aircraft, which are certified by the FAA to operate in the National Airspace System. If
such NASA aircraft are equipped with GPS receivers, they are required to be compliant with
FAA Certification Regulations and are equipped with FAA certified GPS receivers.
Figure K-7: Example of NASA Aircraft Fleet
Moreover, NASA also possesses and operates several Unmanned Aerial Systems (UAS) that are
equipped with GPS receivers. Some of the UAS are designed and developed by NASA Program
Offices, while other UAS are operated under a leasing contract with the UAS developer. UAS
are used by NASA in various manners, from developing UAS
Traffic Management policies and procedures - to performing
airborne science measurements - to performing research and
development of new aircraft materials and aircraft designs.
Regardless what mission or function the UAS is supporting, if
required and necessary for flight in the National Airspace
System, UAS will be equipped with certified aviation
receivers or with general aviation (GAV) receivers.
Figure K-6: NASA
Security Vehicle
Figure K-8: Example of
NASA UAS

13
Spacecraft
Spacecraft, as defined by the International Telecommunication Union (ITU)4
, is a man-made
vehicle which is intended to go beyond the major portion of the Earth's atmosphere. NASA’s
spacecraft portfolio consists of, but not limited to:
• Space vehicles,
• Space stations,
• Space platforms, and
• Satellites.
The orbital mechanics and flight operation of spacecraft, including navigation, POD, metrics
tracking, timing, velocity, and attitude, rely on GNSS signals for accuracy. During development,
spacecraft are typically fitted with either commercially available HP GPS/GNSS receivers or
NASA developed GNSS receivers.
Throughout the years, NASA has developed and continues to develop GNSS receivers that meet
specific mission requirements and designed with the robustness to withstand the harsh elements
of space. Some of these devices are the:
• TurboRogue,
• BlackJack, IGOR, and IGOR+,
• Navigator, and
• TriG.
Employed to perform orbital mechanics and
flight operations, GNSS receivers (commercial
or NASA-developed) are unlikely to be
significantly affected by the ground-based LTE
broadband operations in adjacent bands.
NASA has previously studied the IGOR,
TriG5
, and Navigator6,7
in this mode of
operation. This is due to the configuration and
placement of the antenna. Since most
spacecraft operate within GNSS constellation orbits, antennas are located in the zenith (away
from earth) position of the spacecraft.
Launch Vehicles
Launch vehicles are rockets used to propel a payload from the Earth’s surface to outer space. In
some cases (e.g., sounding rockets), the rockets are designed to carry a scientific measuring
4
International Telecommunication Union, Radiocommunication Sector, (ITU-R), Radio Regulations, Edition 2016,
Volume 1, Chapter I – Terminology and technical characteristics. (http://www.itu.int/pub/R-REG-RR-2016)
5
2011 National Space-Based Positioning, Timing, and Navigation Systems Engineering Forum Report, Subtask 6,
NASA Simulations.
6
GPS Navigator (Nav) Near-band and In-band RFI Susceptibility Report (461-NAV-ANYS-0256), NASA Goddard
Space Flight Center.
7
Limited distribution due to proprietary system information contained within the document.
Figure K-9: Orientation Designations of
Spacecraft

14
device into sub-orbital altitudes; while some rockets are designed with enough inertia and thrust
to enable its payload to entirely escape Earth orbit.
Through the past two decades, the design and development of launch vehicles include the
equipage of GNSS receivers. The use of these receivers facilitates ground control operators by
providing key metric tracking of launch vehicles. Integrated metric tracking units provide
accurate and stable positioning on high dynamic platforms.
More recently, NASA has implemented an Autonomous Flight Safety System (AFSS)8
, which is
a real-time safety system comprised of the ground software used to write mission rules and
convert the mission rules into a mission data load. Coupled with the ground system, the AFSS
includes on-board hardware and software. Specifically, the launch vehicle is equipped with an
Automated Flight Termination Unit (AFTU) used for the Automated Flight Termination System
(AFTS) of the AFSS.
Figure K-10: Antares Rocket Launch, Wallops Island, Oct-2016
(Photo Credit: NASA/Joel Kowsky)
The AFTS augments or replaces the functions of the traditional human-in-the-loop (HITL)
process and procedures. Redundant AFTS processors evaluate data from onboard AFTUs, which
include GNSS receivers and other navigation sensors, and are used to make flight termination
decisions. The mission rules are developed by the local Range Safety Authorities using the
inventory of rule types taken from current HITL operational flight safety practices9
. HP GNSS
receivers are typically configured in the AFTUs to achieve the high-level of accuracy necessary
to track the position of the launch vehicle within the projected launch path safety boundary.
8
An Autonomous Flight Safety System, James B. Bull and Raymond J. Lanzi, NASA Goddard Space Flight Center
and NASA Wallops Flight Facility. https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/20080044860.pdf
9
Reference: Autonomous Flight Termination System Reference Design Hardware, Lisa Valencia, Robert Morrison,
and Roger Zoerner, NASA Kennedy Space Center, FL.
http://www.techbriefs.com/component/content/article/ntb/tech-briefs/machinery-and-automation/24084

15
NASA employs commercial HP GNSS as a part of the AFTU and any such protection criteria
and separation distances afforded to the HP category of receivers under Section 3 will be
applicable to locations where these receivers are used on launch vehicles as part of the AFSS.
Spaceborne Receiver Assessment for Science-Based Applications
NASA has performed an assessment of the potential impacts caused by a proposed terrestrial
LTE network operating in the adjacent band to GPS L1. Two (2) future science missions,
COSMIC-2B and Sentinel-6 (formerly, Jason Continuity of Service (Jason-CS)), were used as
the basis for these assessments. NASA’s assessment is to the TriG receiver performing a science
application using the RO technique.
To determine the impact to the TriG receiver, the aggregate interference power at the output of
the TriG receiver antenna was calculated using MATLAB to model the interference scenario, as
well as the TriG receiver system, and simulate the interference effects to the satellites in orbit.
Satellites operating in LEO gain a much broader view of the earth (dependent upon antenna
characterizations and operating parameters), which must be accounted for in performing the
analysis.
Unlike the assessments performed in Section 3, in-orbit satellites will see a greater number of
potential interference sources (e.g., increased number of terrestrial base stations) and the
aggregate of those interference sources will be the major contributing factor in the assessment,
see Figure K-11.
Figure K-11: Example Satellite View of the U.S. Cities

16
This section describes the modeling and simulation (MS) for a variety of terrestrial LTE base
station deployment scenarios. Further, this section will also describe, where applicable,
assumptions made in the MS, population density of the LTE network, and other dependent
parameters or characteristics. Finally, this section will also provide the results and NASA’s
assessment on impacts/effects on TriG mission performance.
Assumptions
Interference Protection Threshold (TriG)
NASA participated during the DOT ABC Testing of various GPS/GNSS receivers at the ARL
facility in WSMR, NM. One (1) of the various systems NASA tested was the TriG receiver (see
Section 3). The results of the testing produced an interference protection threshold of -73 dBm.
This protection threshold value is based upon the IPC of -1 dB C/No for LTE signals being
present in 1526 – 1536 MHz.
Furthermore, the testing produced a loss-of-lock threshold down to -59 dBm.
Table K-1: TriG Interference Protection Threshold
Parameter Threshold Effect on TriG
-1 dB C/No -73 dBm Degraded performance, inaccurate measurements
Loss-of-Lock -59 to -35 dBm Saturated/jammed (no longer able to receive signals)
Impacted Receiver Satellite Orbit Specifications
The impacted receiver, in the context herein, is referred to at the receiver system that will be
impacted by interference from the interfering source (e.g., terrestrial LTE broadband network).
Typical TriG receiver specifications have been previously described in this Appendix. The
following provides the satellite-specific parameters for each of the assessed Missions.
Table K-2: Simulation Parameters - Satellite Orbit Parameters
Orbit Characteristic COSMIC-2B Sentinel-6
Altitude 800 km 1330 km
Inclination Angle 72° 66°
TriG Receiver Antenna
The TriG receiver system can be configured to use a variety of NASA/JPL developed antennas
to meet its mission needs. The following provides a description of the antenna configurations
used to support COSMIC-2B and Sentinel-6.

17
Antenna Configuration for COSMIC-2B Mission
The antenna configuration to support COSMIC-2B utilizes a set of two (2) proprietary beam
forming array antennas. Each antenna is a 12-element array comprising 3 subarrays of 4
vertically stacked elements with a peak gain at 1530 MHz of +13.4 dBic. Each array uses a 60
cm tall x 40 cm wide backplane mounted on the spacecraft so that the plate is vertical and the
outward normal to the plate is parallel to the spacecraft's velocity vector (assuming circular
orbit). The first antenna is mounted in the forward direction of a satellite (to receive rising
GNSS satellite signals) and the second antenna is mounted in the aft direction (to receive setting
GNSS satellite signals). The TriG receiver has eight (8) independent antenna inputs (three (3)
inputs from each of the two (2) subarrays and two (2) inputs from the antennas performing POD
and space weather data acquisition functions.) Three (3) subarrays (performing the RO
technique) are combined for the fore and aft antennas, increasing the gains by approximately 4.8
dB to a total main beam gain of +18.2 dBic. Note that the gain and beam shape used for the
simulation is from the 4-element subarray. Since each subarray has its own filter/LNA chain, the
effects of RFI apply at the subarray level.
Based on the satellite altitude (for COSMIC-2B altitude = 800 km), the receiver main-beam is
directed towards the earth limb (approximately 26.2° below the satellite velocity vector). Figure
K-12 demonstrates an example of an in-orbit satellite with the forward antenna subarray with its
down-tilt. [Note: For graphical simplicity, the aft subarray is not pictured.] Consequently, the
potential interfering signals from terrestrial LTE BS will be in view of the receive antenna array
main-beams.
Figure K-12: TriG RO Antenna Array Main-Beam Down-tilt (26.2°)
The antenna subarrays are designed to receive right-hand circular polarized signals from the
GNSS satellites. For the analysis, an antenna coupling mismatch (cross-polarization loss) of -3
dB is used (assuming a typical vertically polarized LTE signal).
Figure K-13 and Figure K-14 show the gain pattern for the forward antenna with the main-
beam directed 26.2° below the satellite velocity vector towards earth limb.
Forward Antenna Subarray
Down-tilt Angle
From Satellite velocity
vector : 26.2°

18
Figure K-13: Forward Direction Antenna Array (12-Element, 13.4 dBic @ 1530 MHz,
main-beam pointed towards Earth limb)
Figure K-14: Aft Direction Antenna Array (12-Element, 13.4 dBic @ 1530 MHz, main-
beam pointed towards Earth limb)
Based upon the antenna array specifications and operational parameters, above, the 3 dB antenna
beam width coverage footprint from COSMIC-2B is approximately 1.6 million square miles.
The yellow shaded area over the U.S. in Figure K-15 displays the footprint for the forward
antenna array. [Note: It should be noted that a similar area of coverage (mirror-image in the
horizontal plane) would also be succeeding the satellite.]

19
Figure K-15: COSMIC-2B Antenna 3 dB Beam width Coverage Footprint
COSMIC-2B Antenna Configuration Used for MS Analysis
Although the COSMIC-2B antenna has 2 beamforming arrays the MS was configured to model
only the forward antenna since the aft antenna should show near-identical RFI statistics when
averaged over ten days.
The effects of this modification are discussed in the results section for COSMIC-2B.
Antenna Configuration for Sentinel-6 Mission
Similar to the COSMIC-2B antenna configuration, the antenna configuration to support Sentinel-
6 utilizes a set of two (2) proprietary beam forming array antennas. However, the mission
requirements for Sentinel-6 call for a different array configuration, as well as a difference of
subarrays on the forward and aft directions of the spacecraft.
The forward antenna array is comprised of a six (6) element array in a 2 x 3 configuration. This
array will nominally produce a main beam gain of approximately +15.5 dBic at 1530 MHz.
Based on the orbit altitude of Sentinel-6 (1330 km), the forward antenna is mechanically down-
tilted so that the main-beam is 34.2° below the satellite velocity vector towards earth limb.
The aft antenna array is comprised of a 12-element array in a 4 x 3 configuration. This array will
nominally produce a main beam gain of approximately +17.5 dBic at 1530 MHz. Based on the
orbit altitude of Sentinel-6 (1330 km), the aft antenna is electrically phased down (down-tilted)
by 22°, as well as mechanically down-tilted an additional 12.0° below the satellite velocity
vector towards earth limb.

20
Sentinel-6 Antenna Configuration Used for MS Analysis
The Sentinel-6 antenna will digitally combine the outputs of each of the three subarrays on the
fore and aft antennas. Because each RF front end is separate for each subarray, the effect on a
single subarray was analyzed for degradation and saturation. For the forward subarray, a 2-
element array (2 x 1 configuration) with a peak main beam gain of +10.5 dBic at 1530 MHz was
modeled for the simulation. In the aft subarray, a 4-element array (4 x 1 configuration) with a
peak main beam gain of +12.5 dBic at 1530 MHz was modeled for the simulation.
The effects of this modification are discussed in the results section for Sentinel-6.
Summary of TriG Receiver System Characteristics Used for Analyses
Table K-3 summarizes the satellite TriG receiver system characteristics for the analyses
performed on COSMIC-2B and Sentinel-6. The interference threshold in this table is the RFI
power at the output of the flight RO antenna which causes a -1 dB C/No degradation in the TriG
receiver as used in the COSMIC2 mission. It was derived from the power density observed by
the 0 dBiL standard gain horn used in during the DOT ABC test at a RFI power level causing a 1
dB C/No degradation. Since the TriG choke ring antenna was located at a different spot, it
actually received about 3.2 dB more RFI power per meter squared (m2
). In addition, the choke
ring antenna had about +3.7 dBi linear gain toward the RFI source, adding 3.7 dB to the
threshold power. After these corrections, the LTE power at 1530 MHz that causes a 1 dB C/No
degradation is -78.2 dBm + 3.2 dB + 3.7 dB = -71.3 dBm, defined at the output of the receive
antenna.
Another adjustment that was made to estimate the effect on the flight receiver is the difference in
noise floors due to the extra antenna temperature from black body radiation coming from the
ceiling and walls of the WSMR anechoic chamber. During the test, the noise floor is estimated
to be 349 Kelvin (K). This is based on preamplifier (Preamp) noise of 51 K, antenna temp of 300
K, and filter loss of 0.8 dB. The noise floor in flight is estimated to be 224 K based on Preamp
noise of 51 K, antenna temp of 150 K, and filter loss of 0.8 dB. This difference shows an
adjustment to lower the 1 dB threshold by 1.9 dB. Therefore, the normalized in-flight RFI power
of is calculated to be approximately -73 dBm (-71.3 dBm – 1.9 dB = -73.2 dBm) from the
antenna corresponding to a -1 dB degradation of C/No.
Table K-3: Summary Table of Satellite TriG Receiver Characteristics Used for MS
Receiver Characteristic COSMIC-2B Sentinel-6
Satellite Orbit Altitude 800 km 1330 km
Satellite Orbit Inclination Angle 72° 66°
TriG Forward Receive Antenna Type 12-Element Array 6-Element Array
TriG Forward Receive Antenna
Downtilt (relative to satellite velocity
vector)
26.2° 34.2°

21
TriG Forward Receiver Antenna
Subarray Gain @ 1530 MHz
+ 13.4 dBic + 10.5 dBic
TriG Aft Receive Antenna Type Not modeled 12-Element Array
TriG Aft Receive Antenna Downtilt
(relative to satellite velocity vector)
Not modeled 34.0°
TriG Aft Receiver Antenna Subarray
Gain @ 1530 MHz
Not modeled + 12.5 dBic
Interference Threshold (-1 db C/No) - 73 dBm - 73 dBm
Loss-of-Lock (Note: The LOL value
ranged from a low of -59 dBm for Test
04 at 1525 MHz to a high of -35 dBm
for Test 04 at 1530 MHz, all corrected
for antenna location and gain.)
- 59 to -35 dBm - 59 to -35 dBm
Antenna Coupling Mismatch (Cross-
Polarization Loss)
- 3 dB - 3 dB
Terrestrial LTE Deployment Scenarios
The aggregate interference is dependent upon several factors. A few of those factors are the
satellite related, to include, orbital parameters and receiver system characteristics. The other
determining factor comes from the interference sources. Most importantly, the transmitter
characteristics and the total number of sources (e.g., LTE BS). Since TriG receiver systems
(performing the RO technique) operate in LEO, they have a direct line-of-sight to a broad area of
the U.S., and the aggregate interference is dependent upon the long-term deployment scenario of
the LTE operator.
The following describes the LTE parameters and the developed scenarios used during MS.
Base Stations used for LTE Deployment
To model the terrestrial LTE base station deployment, the analyses used base station (BS) macro
and microcell parameters, which are primarily derived from ITU-R M.2292.
For this analysis, the following BS macro and microcell antenna bore-sites with respect to True
North were assumed:
• Macrocell Sector-1 bore-site: 0°
• Microcell Sector bore-site: Randomly selected from (0°, 120°, 240°)

22
NASA used two (2) different methodologies to determine the total number of BS that could be
deployed to support the LTE network. The assumptions used for each of the methodologies are
described below and resulted in a different number of cell sites.
City Zone Model
The City Zone model was used to determine the physical area around a city center location that
the simulated LTE network would be deployed over. The baseline City Zone model was chosen
to conform to the only available accepted model given in ITU Report ITU-R SA.2325-010
(International Mobile Telecommunication (IMT) sharing at 2GHz) for an BS deployment based
on three (3) zones (e.g., urban, suburban, and rural) with given radial distances from a city center
latitude/longitude location. Figure K-16 demonstrates an example of the City Zone model with
the typical macro cellular hexagonal grid layout deployed about a city center.
Because the LTE services to be provided by the proposed and analyzed network may not be as
widespread in terms of city area as the conventional LTE deployment described in SA.2325-0 a
second City Zone model with a smaller Suburban and Rural zone size was analyzed. Parameters
for both the City Zone models are listed in Table K-4.
Figure K-16: Base Station Deployment Zone Model (Report ITU-R SA. 2325-0)
10
Reference, https://www.itu.int/dms_pub/itu-r/opb/rep/R-REP-SA.2325-2014-PDF-E.pdf

23
Table K-4: Zone Model - BS Zone-specific Radial Distance from City Center
Zone Model
Urban Zone
(km)
Suburban Zone
(km)
Rural Zone (km)
1 0 – 3 3 – 20 20 – 50
2 0 – 3 3 – 10 10 – 30
City Population Size / Base Station Cell Radius
In addition to a City Zone model it was necessary to define the BS cell radius (CR) parameter in
order to determine the BS grid layout within each City Zone. The typical ITU-R M.2292 zone
values listed in Table K-6 were used as the baseline cell radius (CR) in the simulation.
Table K-5: Typical Cell Radius (CR) - ITU-R M.2292
Zone type City Population CR (km)
Urban All 0.5
Suburban All 1.0
Rural All 5.0
In consideration to the where the proposed LTE network is to be deployed, the size of the city
population was an additional parameter that was included in the simulations. If a U.S. city had a
population of greater than 125,000, but less than 250,000, it was included in the analyses for half
of the simulations. Cities with populations of over 250,000 were included in all simulations.
Accordingly, the number of assumed cities included in each simulation was chosen from:
• City Population 125K: 225 cities or
• City Population 250K: 82 cities
Additionally, since a smaller population city could have a smaller amount of Base Stations with a
larger Cell Radius (CR), then the typical M.2292 CR values where scaled by the city population
and included in the set of simulation runs. Table K-7 shows the addition inclusion of the largest
M.2292 CR Table values.

24
Table K-7: Cell Radius Scaled by City Population Density (ITU-R M.2292)
Zone type City Population (in 1000s) CR (km)
Urban 125 POP 250
250 POP 500
POP 500
1.0
0.75
0.5 (Typical)
Suburban 125 POP 250
250 POP 500
POP 500
2.0
1.5
1.0 (Typical)
Rural 125 POP 250
250 POP 500
POP 500
10.0
10.0
5.0 (Typical)
Total Number of Base Stations in Simulations
Using the set of Zone Model, City Population and Cell Radius parameters, NASA calculated the
total number of BS required for deployment for each simulation run. Table K-8 depicts the
number of base stations for the set of three parameters for a LTE network deployment consisting
of only macrocells. Table K-9 accounts for microcells to be included in the LTE network
deployment.
Table K-8: Total # of BS (Macrocell Deployment Only)
Zone Model
City
Population
(in 1000s)
Cell Radius
Number of BS
Urban Suburban Rural Total
1 125 Table K.6 11,700 143,100 29,700 184,500
1 250 Table K.6 4,264 52,152 10,824 67,240
1 125 Table K.7 5,330 58,962 10,320 74,612
1 250 Table K.7 3,024 35,796 6030 44,868
2 125 Table K.6 11,700 33,750 12,150 57,600
2 250 Table K.6 4,264 12,300 4,428 20,992
2 125 Table K.7 5,330 13,500 5,310 24,140
2 250 Table K.7 3,042 8352 2,736 14,130

25
Table K-9: Total # of BS (Macro + Microcells)
Zone Model
City
Population
(in 1000s)
Cell
Radius
Number of BS
Macrocells Microcells Total
1 125 Table K.6 184,500 97,686 282,186
1 250 Table K.6 67,240 35,601 102,841
1 125 Table K.7 74,612 41,014 115,626
1 250 Table K.7 44,868 24,609 69,477
2 125 Table K.6 57,600 36,450 94,050
2 250 Table K.6 20,992 13,284 34,276
2 125 Table K.7 24,140 15,555 39,695
2 250 Table K.7 14,130 9,240 23,370
Additional LTE Network Deployment Assumptions for Analysis
In addition to the parameters described above, the following simulation parameters were
considered and chosen by NASA for the analysis performed.
• Since specific latitude and longitude locations for the BS in each city were not available,
BS are placed at respective city center latitude/longitude and BS power aggregated for
urban, suburban and rural BS transmitters to get single equivalent urban, suburban, rural
and microcell BS.
Rationale 1: The angular separation between two (2) BS separated by 10 km is only 0.7°,
assuming a TriG receiver at 800 km altitude. This angular separation is relatively small
with respect to the transmitter and receiver antenna gain patterns.
Rational 2: The time and resources required to model separate BS locations for each city
would be exhaustive. Further, the computational time to run the simulations and amount
of processing power would be extensive.
• BS antenna side-lobe pattern:
o ITU-R F.1336-4 Recommends 3.1. (Macro)
o ITU-R F.1336-4 Recommends 3.2. (Micro)
• Per M.2292, 30% of the macrocell BS are below rooftop and the simulation considered
half of the 30% blocked from contributing interference and have already been excluded
in the total BS calculations in Table K-8.

26
• Per M.2292, microcell BS antennas are below rooftop with 50% of the microcells in the
urban zone and 30% of the microcells in the suburban zone considered blocked. These
BS have already been excluded in the total BS calculations in Table K-9.
• Elevation Mask:
Consideration given to blockage from terrain, vegetation, and addition man-made
structures. This was simulated by providing a 5° transmitter elevation mask in the
vertical plane of the transmitter, 360° around the BS in the horizontal plane.
Two (2) BS mask angles are utilized for the analysis:
o A 0° elevation mask on the BS so that all BS which see the satellite above 0°
elevation angle are included in the aggregate interference calculation, and
o A 5° mask angle so that only BS which see the satellite above 5° elevation angle
contribute to the aggregate interference.
• One (1) 10 MHz LTE channel per sector.
• Propagation Loss: Free-space
• BS Activity Factor (AF):
An AF of 3 dB, corresponding to 50% of the base stations transmitting simultaneously, is
used throughout the analysis.
Note: If 100% of the base stations are transmitting simultaneously, the peak interference
levels in the results will be 3 dB higher. This will also hold true for other resultant
statistics, as well.
• BS Transmitter Power (EIRP):
Table K-10 depicts the nominal transmit power used for some of the simulations (as per
ITU-R M.2292). Considerations were also given to the maximum transmit powers of
+10 dBW11
and +32 dBW12
EIRP per channel per sector.
11
Proposed maximum transmit power per channel per sector through negotiations with the FAA.
12
Maximum transmit power per channel per sector as authorized in the current FCC license.

27
Table K-10: Assumed Maximum Transmitter Levels per Sector
(Typical per ITU-R M.2292)
BS Type
Typical Max. Transmit
Power/Channel/Sector (EIRP)
Macrocell - Urban 26 dBW
Macrocell - Suburban 26 dBW
Macrocell - Rural 28 dBW
Microcell (any zone) 7 dBW
TriG Receiver Analysis
Two (2) NASA missions (COSMIC-2B and Sentinel-6) that include the TriG receiver, as a
science-based function (e.g., RO technique) were utilized for analysis. A MATLAB simulation
program was developed to model the receiver on-board a satellite, using mission-specific
parameters, and interference statistics were calculated for an LTE network deployment of BS
distributed in U.S. cities.
MATLAB Simulation
For the spaceborne receiver analysis the aggregate interference power at the output of the GPS
receiver antenna is calculated at ten (10) second time steps in the satellite orbit from BS
distributed among U.S. cities. The MATLAB program was setup to model a 10-day orbit of the
satellite. Figure K-17 provides an example of the COSMIC-2B satellite simulation of a 10-day
orbit.
Figure K-17: Ground Track of COSMIC-2B Orbital Path (10-Day Simulation at 10-Second
Time Steps)
A similar program written in Python was implemented for the Sentinel-6 simulation.

28
Aggregate Interference Calculation
The analysis calculates the Io/No value and is not dependent upon the carrier (C) signal.
Thresholds for determining the saturation (-1 dB C/No) and jammed (loss-of-lock) values of the
TriG are discussed in this report.
The aggregate interference to the receiver antenna output is calculated using a summation of the
interference from each source. A simple link budget formula is used to calculate the interference
received by a single source, LTE BS. The total aggregate interference is determined through the
summation of interference from the individual sources, see Equation K-2.
Rx Int Pwragg = ∑(Int sources) Tx Pwr (EIRP)off-boresite – FSPL – Pol Loss + Rx Ant Gainoff-
boresite
Equation K-2: Formula in Determining Received Aggregate Interference at the TriG
Antenna Output
Where,
Rx Int Pwragg = Aggregate interference power level (dBm)
Tx Pwr (EIRP) off-boresite = Tx power output including antenna off-boresite
calculations (dBm) (See below)
FSPL = Free Space Path Loss (dB)
Pol Loss = Loss of dissimilar polarizations (Linear to RCHP Polarization = - 3
dB)
Rx Ant Gain off-boresite = Rx antenna gain including antenna off-boresite
calculations (dBic)
The BS sector antenna gain towards the satellite is calculated by first determining the appropriate
azimuth (AZ) (horizontal plane) and elevation (EL) (vertical plane) angles based on the BS and
satellite geometry. The antenna off-boresite gain is calculated by, first, summing the AZ plane
discrimination with the EL plane discrimination and, secondly, subtracting this total
discrimination from the maximum sector gain13
to get the net sector gain towards the satellite.
Note: The maximum interference from an BS will occur when it sees the satellite at low
elevation angles.
Simulation Runs
A total of 96 simulation runs were performed for COSMIC-2B, while a lesser number, but still
representative, number of runs (16 runs) were performed for Sentinel-6. Each of the simulation
runs varied one or more LTE BS deployment parameters.
While it is unknown for how the LTE operator will be performing their network deployment, the
variations in simulation runs should be demonstrative. Further, the variations in runs may be
representative of an LTE network through its various phases of deployment (initial deployment
through full deployment). Table K-6 summarizes the various simulation runs.
13
As defined in ITU-R M.2292 and ITU.R F.1336-4.

29
Table K-6: Summary of Simulation Runs
Run
Sim
No.
Run
Designator
COSMIC-
2B
Sentinel-6 BS Tx Power
Zone
Model
City
Population
Cell
Radius
Elevation
Mask
Macrocell
Only
Macro +
Microcell
Total # of
Base
Stations
1 1 a X M.2292 levels 1 125K Typical 0° X 184,500
2 1 b X X M.2292 levels 1 125K Typical 5° X 184,500
3 1 c X 32 dBW 1 125K Typical 0° X 184,500
4 1 d X 32 dBW 1 125K Typical 5° X 184,500
5 1 e X 10 dBW 1 125K Typical 0° X 184,500
6 1 f X X 10 dBW 1 125K Typical 5° X 184,500
8 2 b X X M.2292 levels 1 250K Typical 5° X 67,240
9 2 c X 32 dBW 1 250K Typical 0° X 67,240
10 2 d X 32 dBW 1 250K Typical 5° X 67,240
11 2 e X 10 dBW 1 250K Typical 0° X 67,240
12 2 f X X 10 dBW 1 250K Typical 5° X 67,240
13 3 a X X M.2292 levels 1 125K Scaled 0° X 74,612
14 3 b X X M.2292 levels 1 125K Scaled 5° X 74,612
15 3 c X X 32 dBW 1 125K Scaled 0° X 74,612
16 3 d X X 32 dBW 1 125K Scaled 5° X 74,612
17 3 e X X 10 dBW 1 125K Scaled 0° X 74,612
18 3 f X X 10 dBW 1 125K Scaled 5° X 74,612

30
19 4 a X X M.2292 levels 1 250K Scaled 0° X 44,850
20 4 b X X M.2292 levels 1 250K Scaled 5° X 44,850
21 4 c X X 32 dBW 1 250K Scaled 0° X 44,850
22 4 d X X 32 dBW 1 250K Scaled 5° X 44,850
23 4 e X X 10 dBW 1 250K Scaled 0° X 44,850
24 4 f X X 10 dBW 1 250K Scaled 5° X 44,850
26 5 b X M.2292 levels 2 125K Typical 5° X 57,600
27 5 c X 32 dBW 2 125K Typical 0° X 57,600
28 5 d X 32 dBW 2 125K Typical 5° X 57,600
29 5 e X 10 dBW 2 125K Typical 0° X 57,600
30 5 f X 10 dBW 2 125K Typical 5° X 57,600
33 6 c X 32 dBW 2 250K Typical 0° X 20,992
34 6 d X 32 dBW 2 250K Typical 5° X 20,992
35 6 e X 10 dBW 2 250K Typical 0° X 20,992
36 6 f X 10 dBW 2 250K Typical 5° X 20,992
37 7 a X M.2292 levels 2 125K Scaled 0° X 24,140
38 7 b X M.2292 levels 2 125K Scaled 5° X 24,140

31
39 7 c X 32 dBW 2 125K Scaled 0° X 24,140
40 7 d X 32 dBW 2 125K Scaled 5° X 24,140
41 7 e X 10 dBW 2 125K Scaled 0° X 24,140
42 7 f X 10 dBW 2 125K Scaled 5° X 24,140
45 8 c X 32 dBW 2 250K Scaled 0° X 14,130
46 8 d X 32 dBW 2 250K Scaled 5° X 14,130
47 8 e X 10 dBW 2 250K Scaled 0° X 14,130
48 8 f X 10 dBW 2 250K Scaled 5° X 14,130
51 9 c X 32 dBW 1 125K Typical 0° X 282,186
52 9 d X 32 dBW 1 125K Typical 5° X 282,186
53 9 e X 10 dBW 1 125K Typical 0° X 282,186
54 9 f X 10 dBW 1 125K Typical 5° X 282,186
57 10 c X 32 dBW 1 250K Typical 0° X 102,841
58 10 d X 32 dBW 1 250K Typical 5° X 102,841

32
59 10 e X 10 dBW 1 250K Typical 0° X 102,841
60 10 f X 10 dBW 1 250K Typical 5° X 102,841
63 11 c X 32 dBW 1 125K Scaled 0° X 115,626
64 11 d X 32 dBW 1 125K Scaled 5° X 115,626
65 11 e X 10 dBW 1 125K Scaled 0° X 115,626
66 11 f X 10 dBW 1 125K Scaled 5° X 115,626
69 12 c X 32 dBW 1 250K Scaled 0° X 69,477
70 12 d X 32 dBW 1 250K Scaled 5° X 69,477
71 12 e X 10 dBW 1 250K Scaled 0° X 69,477
72 12 f X 10 dBW 1 250K Scaled 5° X 69,477
75 13 c X 32 dBW 2 125K Typical 0° X 94,050
76 13 d X 32 dBW 2 125K Typical 5° X 94,050
77 13 e X 10 dBW 2 125K Typical 0° X 94,050
78 13 f X 10 dBW 2 125K Typical 5° X 94,050

33
81 14 c X 32 dBW 2 250K Typical 0° X 34,276
82 14 d X 32 dBW 2 250K Typical 5° X 34,276
83 14 e X 10 dBW 2 250K Typical 0° X 34,276
84 14 f X 10 dBW 2 250K Typical 5° X 34,276
87 15 c X 32 dBW 2 125K Scaled 0° X 39,695
88 15 d X 32 dBW 2 125K Scaled 5° X 39,695
89 15 e X 10 dBW 2 125K Scaled 0° X 39,695
90 15 f X 10 dBW 2 125K Scaled 5° X 39,695
93 16 c X 32 dBW 2 250K Scaled 0° X 23,370
94 16 d X 32 dBW 2 250K Scaled 5° X 23,370
95 16 e X 10 dBW 2 250K Scaled 0° X 23,370
96 16 f X 10 dBW 2 250K Scaled 5° X 23,370

34
Results
The aggregate interference results for the TriG receiver, functioning as a science measurement
instrument, are presented in the following sections.
The received aggregate interference levels calculated during the simulations range from -90 dBm
to -40 dBm.
The following tables use an aggregate interference threshold of -73 dBm (1526 – 1536 MHz)
which corresponds to a -1 dB degradation of receiver C/No.
It should be noted that the loss-of-lock threshold for the TriG receiver occurs between -59 to -35
dBm aggregate interference power in the 1526-1536 MHz band. Loss-of-Lock at -59 dBm was
seen in Test 04 with RFI at 1525 MHz and LOL at -35 dBm was seen in Test 04 at 1530 MHz.
The entries in the results tables are interpreted as follows:
• Column 3: Max Int. Level (dBm)
Indicates the maximum aggregate interference level calculated at the receiver antenna
output.
Note: Any value ≥ -66 dBm in this column indicates that there is sufficient aggregate
interference received from the terrestrial LTE network for the TriG receiver to lose lock.
• Column 4: % Time Threshold
Indicates the percent time, over the 10-day simulation period, where the aggregate
interference at the TriG receiver antenna output exceeds the threshold level (-73 dBm).
As an example, if the value is about 10% of the time, the TriG receiver will have C/No
degraded by at least 1 dB for a cumulative of 24 hours. This is calculated by, as an
example:
10 days (total period of simulation run) = 240 hours
% Time Threshold = 10%
10% of 10 days (240 hours) = 0.10 x 240 = 24 hours
Note: The value reported represents the % Time Threshold for the entire 10-days of the
simulation, to include the time and instances where the continental U.S. is not within the
field-of-view of the satellite. Consideration must be taken based on this. If the master
time schedule only included the instances where the continental U.S. (and surrounding
bodies of water) were in the field-of-view of the satellite, these values would increase.

35
• Column 5: # of Int Events
Indicates that over the 10-day period, the total number of interference events which
exceed the -73 dBm threshold.
Note: The interference time intervals for each interference occurrence may be short or
long depending on how many interfering BS the satellite sees on the particular orbit pass
over the U.S. The sum duration of all of the interference events (provided in this column)
is the reported in column 4 (% Time Threshold). Furthermore, it should be noted that
there can be multiple interference events for a single satellite pass, as different BS pass
through the field-of-view of the TriG receiver antenna.
• Column 6: Avg Dur Int Event (min)
Indicates the mean average duration (in minutes) of an interference event for the entire
10-day period.
Note: As discussed in Section 5.3.1, the duration of an atmospheric occultation (as the
signal path moves from skimming the Earth‘s surface to an altitude of about 100 km) is
one to two minutes.
• Column 7: Max Int Event (min)
Indicates the maximum duration (in minutes) that was recorded for a single interference
event over the 10-day period.
• Column 8: Max Allow EIRP Level (dBW)
Indicates a reverse-engineered maximum BS transmitter power level (in dBW)
distributed across a 10 MHz bandwidth per channel per sector. The calculated level is
based on the maximum interference level received during the 10-day period.
Note 1: The reverse-engineered value calculated in this column would bring the
interference level below the -73 dBm threshold value. However, it should be also noted
that interference to the TriG receiver occurs well before the -73 dBm threshold value
occurs, which causes degradation in scientific measurements (e.g., interference occurs at
interference levels -90 dBm to -73 dBm (threshold).
Note 2: Where applicable (i.e., simulations that utilized variable maximum transmitter
power levels), the maximum allowable EIRP level is linearly calculated for each zonal
category of BS sector.
As an example, if the maximum interference level (column 3) indicates -70 dBm, the BS
transmitter power needs to be reduced by 3.1 dBm in order for the received interference
to be below the -73 dBm threshold. The 3.1 dBm reduction in power is linearly attributed
to each of the maximum transmitter power for the urban/suburban (+26 dBW), rural (+28

36
dBW), and microcells (+7 dBW). The resulting maximum allowable transmitter power is
calculated for the urban/suburban zone as +22.9 dBW, rural zone as +24.9 dBW, and
microcells as +3.9 dBW.
Results Caveats
Caveat 1:
The results are only for the LTE deployment scenarios derived from parameters outlined.
Deviation of such LTE system characteristics from ITU-R M.2292 may adversely impact the
interference received at the satellite. This is especially true if the typical BS antennas vary in the
vertical plane from what was defined in ITU-R F.1336-4, or if the nominal down-tilt angles, as
defined in ITU-R M.2292, are deployed at 0 deg or with an up-tilt (e.g., more LTE BS signal
energy pointing directly over the horizon or into the atmosphere).
Caveat 2:
The results presented in the following sections are intended to draw no conclusions or make any
recommendations as to what level of interference may be tolerated by the other missions
employing the TriG receiver for science applications. Aggregate interference received by the
TriG receiver system in-orbit is dependent upon the satellite orbit parameters and receive
antenna configurations.
Caveat 3:
The results are for the simulated operational use of the TriG receiver while in-orbit. It should be
noted that the TriG receivers are currently researched, developed and tested (RDT) in facilities
that are not electromagnetically shielded from the existing RF environment. As such, the TriG
receivers may be impacted by LTE BS sites located within close proximity. The effects to the
RDT facilities have not been studied and additional analyses would be required to further
understand the impacts to the TriG receivers at the RDT facilities.
Results for COSMIC-2B
Tables Table K-7 through

37
Table K-12 provides a results summary of the analyses performed for the TriG receiver
simulated aboard a single COSMIC-2B satellite, and for all simulation parameters shown in
Table K-6. (The COSMIC-2B mission is comprised of six (6) total satellites.)
[Note: Only the forward antenna array was used in the MS. In reality, COSMIC-2B will utilize
a set of 3 subarrays on the forward and aft ends of the satellite. These three (3) subarrays (each
having +13.4 dBic gain at 1530 MHz) will be digitally combined in the TriG receiver to achieve
a total of +16.7 dBic antenna gain at 1530 MHz in both the fore and aft antennas.]
The results below show results for the fore antenna. The viewing geometries will be similar for
the aft antenna, and the calculated values for the tables below would be expected to have similar
values from the aft antenna. There will be slight variations due to the fact the individual Base
Stations are viewed from azimuths which differ by about 180 degrees (in the BS reference
frame).
Table K-7: COSMIC-2B Interference Results
(Macro BS Only, Urban/Suburban: Tx Power +26 dBW,
Rural Tx Power: +28 dBW)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
# of Int
Events
Avg Dur
Int.
Event
(min)
Max Int
Event
(min)
Max Allow EIRP Level
(dBW)
Urban/
Suburban
BS
Rural
BS
1 a -57 3.3 83 5.5 11.0 10 12
1 b -62 2.1 61 4.8 8.8 15 17
2 a -62 1.8 59 4.1 9.0 15 17
2 b -67 1.1 43 3.4 6.3 20 22
3 a -62 1.9 62 4.3 9.0 15 17
3 b -66 1.2 40 4.0 7.0 19 21
4 a -64 1.3 44 3.9 7.7 17 19
4 b -68 0.6 31 2.8 4.8 21 23
5 a -63 1.6 52 4.2 8.5 16 18
5 b -67 0.9 32 3.7 6.7 20 22
6 a -68 0.7 43 2.2 5.5 21 23
6 b -72 0.1 10 1.9 2.7 25 27
7 a -67 0.8 32 3.4 6.0 20 22
7 b -71 0.2 19 1.4 3.0 24 26
8 a -69 0.4 31 1.6 4.2 22 24

38
8 b -74 0.0 0 0.0 0.0 27 29

39
(Macro BS Only, All BS Tx Power +32 dBW)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
# of Int
Events
Avg Dur
Int. Event
(min)
Max Int
Event
(min)
Max Allow
EIRP Level
(dBW)
1 c -52 5.4 137 5.5 13.2 11
1 d -56 3.7 84 6.2 10.7 15
2 c -57 3.7 115 4.5 10.7 16
2 d -61 2.3 67 4.9 9.3 20
3 c -57 3.9 93 5.8 11.5 16
3 d -61 2.5 70 5.1 9.5 20
4 c -58 3.1 87 5.0 10.5 17
4 d -63 2.0 57 4.8 8.0 22
5 c -57 3.4 83 5.7 11.2 16
5 d -61 2.2 57 5.4 8.8 20
6 c -62 1.9 62 4.2 9.2 21
6 d -66 1.2 47 3.5 6.5 25
7 c -62 2.1 66 4.4 9.2 21
7 d -66 1.3 47 3.9 7.3 25
8 c -64 1.4 44 4.3 7.8 23
8 d -68 0.7 32 3.2 5.3 27
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
# of Int
Events
Avg Dur
Int. Event
(min)
Max Int
Event
(min)
Max Allow
EIRP Level
(dBW)14
1 e -74 0.0 0 0.0 0.0 10
1 f -78 0.0 0 0.0 0.0 10
2 e -79 0.0 0 0.0 0.0 10
2 f -83 0.0 0 0.0 0.0 10
3 e -79 0.0 0 0.0 0.0 10
14
Based on the assumption that the maximum transmitter power level is limited to +10 dBW.

40
3 f -83 0.0 0 0.0 0.0 10
4 e -80 0.0 0 0.0 0.0 10
4 f -85 0.0 0 0.0 0.0 10
5 e -79 0.0 0 0.0 0.0 10
5 f -83 0.0 0 0.0 0.0 10
6 e -84 0.0 0 0.0 0.0 10
6 f -88 0.0 0 0.0 0.0 10
7 e -84 0.0 0 0.0 0.0 10
7 f -88 0.0 0 0.0 0.0 10
8 e -86 0.0 0 0.0 0.0 10
8 f -90 0.0 0 0.0 0.0 10
(Macro + Microcells, Urban/Suburban: Tx Power +26 dBW,
Rural Tx Power: +28 dBW, Microcell Tx Power +7 dBW)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
%
Time
Thresh
# of
Int
Events
Avg Dur
Int. Event
(min)
Max Int
Event
(min)
Max Allow EIRP Level (dBW)
Urban/
Suburban
BS
Rural
BS
Microcell
9 a -57 3.3 81 5.8 11.2 10 12 -9
9 b -62 2.1 60 5.0 8.8 15 17 -4
10 a -62 1.8 59 4.3 9.0 15 17 -4
10 b -66 1.1 42 3.8 6.3 19 21 0
11 a -62 2.0 62 4.5 9.0 15 17 -4
11 b -66 1.2 43 3.9 7.0 19 21 0
12 a -64 1.3 44 4.1 7.8 17 19 -2
12 b -68 0.7 30 3.1 5.0 21 23 2
13 a -63 1.7 52 4.5 8.7 16 18 -3
13 b -67 1.0 43 3.1 6.8 20 22 1
14 a -68 0.7 40 2.5 5.5 21 23 2
14 b -71 0.2 15 1.4 2.8 24 26 5
15 a -67 0.8 33 3.4 6.2 20 22 1

41
15 b -71 0.2 18 1.8 3.0 24 26 5
16 a -69 0.4 35 1.5 4.3 22 24 3
16 b -73 0.0 4 0.1 0.2 26 28 7
(Macro + Microcells, All S Tx Power +32 dBW/10MHz)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
%
Time
Thresh
# of
Int
Events
Avg Dur
Int. Event
(min)
Max Int
Event
(min)
Max Allow
EIRP Level
(dBW)
9 c -49 8.5 160 7.5 15.5 8
9 d -50 6.3 136 6.5 13.5 9
10 c -53 7.0 147 6.7 13.5 12
10 d -54 5.2 128 5.6 11.5 13
11 c -53 6.9 137 7.0 14.3 12
11 d -54 5.1 111 6.5 11.8 13
12 c -55 6.4 135 6.7 13.2 14
12 d -56 4.6 132 4.9 11.2 15
13 c -53 6.5 128 7.2 13.8 12
13 d -54 4.9 111 6.2 11.7 13
14 c -58 4.9 119 5.7 11.8 17
14 d -59 3.5 97 5.0 10.2 18
15 c -57 4.8 95 7.1 12.3 16
15 d -58 3.5 86 5.7 10.5 17
16 c -59 3.8 91 5.9 11.0 18
16 d -61 2.7 71 5.3 9.5 20

42
(Macro + Microcells, All BS Tx Power +10 dBW)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
%
Time
Thresh
# of
Int
Events
Avg Dur
Int. Event
(min)
Max Int
Event
(min)
Max Allow
EIRP Level
(dBW)15
9 e -71 0.5 32 1.9 3.8 7
9 f -72 0.1 12 1.0 2.2 8
10 e -75 0.0 0 0.0 0.0 10
10 f -76 0.0 0 0.0 0.0 10
11 e -75 0.0 0 0.0 0.0 10
11 f -76 0.0 0 0.0 0.0 10
12 e -77 0.0 0 0.0 0.0 10
12 f -78 0.0 0 0.0 0.0 10
13 e -75 0.0 0 0.0 0.0 10
13 f -76 0.0 0 0.0 0.0 10
14 e -80 0.0 0 0.0 0.0 10
14 f -81 0.0 0 0.0 0.0 10
15 e -79 0.0 0 0.0 0.0 10
15 f -80 0.0 0 0.0 0.0 10
16 e -81 0.0 0 0.0 0.0 10
16 f -83 0.0 0 0.0 0.0 10
Results for Sentinel-6
Tables Table K-13 through Table K-15 report the results of the analyses performed for the TriG
receiver simulated aboard the Sentinel-6 satellite.
[Note: A 2-element subarray (2 x 1 configuration) with a gain of +10.5 dBic at 1530 MHz was
modeled for the forward subarray, and a 4-element subarray (4 x 1 configuration) with a peak
gain of +12.5 dBic at 1530 MHz was modeled for the aft subarray. Results represent the total
interference in both antennas. When the full array output is combined, the 6-element forward
array consisting of three 2x1 subarrays will have a main beam gain of +15.5 dBic at 1530 MHz,
and the 12-element aft array consisting of three 4x1 subarrays will have a main beam gain of
+17.5 dBic at 1530 MHz.]
15

43
Table K-13: Sentinel-6 Interference Results
(Macro BS Only, Urban/Suburban: Tx Power +26 dBW,
Rural Tx Power: +28 dBW)
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
Max Allow EIRP Level
(dBW)
Urban/
Suburban
BS
Rural BS
1 b -66 3.7 19 21
2 b -70 1.9 23 25
3 b -72 0.6 25 27
4 b -76 0 29 31
3 a -70 2.1 23 25
4 a -74 0.2 27 29
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
Max Allow
EIRP Level
(dBW)
3 c -64 5.8 23
3 d -66 3.7 25
4 c -68 3.4 27
4 d -70 1.9 29

44
Sim
No.
Run
Designator
Max int.
Level
(dBm)
% Time
Thresh
Max Allow
EIRP Level
(dBW)16
1 f -76 0 10
2 f -76 0 10
3 f -78 0 10
4 f -82 0 10
3 e -76 0 10
4 e -80 0 10
Results Summary
The results tables represent a myriad of LTE BS deployment scenarios and reports the maximum
allowable EIRP levels for the terrestrial LTE BS sectors per channel.
In the case of COSMIC-2B, for the simple scenario of macro cell BS at 32 dBW EIRP, as the
number of stations decreases from simulation 1 to 2 for the zone-1 model, and from simulation 5
to 6 for the zone-2 model, there is about 5 dB less interference in zone-2 compared to zone-1,
which is expected because the zone-2 model uses about 3 times less stations. There is about 4
dB less interference in models using transmitter elevation mask of 5o
(run d) compared to the 0o
mask (run c), indicating that less than half of the available stations affect the satellite in the 5o
mask case.
For the most challenging model (1c), using 184,500 macro cell stations, the tolerable EIRP is 11
dBW(i.e., 12.6 Watts). For a deployment of macro and microcells, utilizing the same transmitter
power, the maximum tolerable EIRP is approximately 8 dBW(i.e., 6.3 Watts).
In the case of the Sentinel-6, for the simple scenario of macro cell BS at 32 dBW EIRP, as the
number of stations decreases from simulation 3 to 4 for the zone-1 model, there is about 2 dB
less interference in models using transmitter elevation mask of 5o
(run d) compared to the 0o
mask (run c). For the most challenging model (3c), using 74,612 macro cell stations, the
tolerable EIRP is 23 dBW.
It should be noted that simulations 3 and 4 use the aforementioned variation of the cell model,
referred to as ‘scaled’ model, in which the cell radius increases up to double its typical value, as
the city population decreases; this results in fewer stations, and less interference, compared to the
simulations 1 and 2.
16

45
These tolerances only predict the impact to two (2) NASA Missions (COSMIC-2B and Sentinel-
6), and the results from these simulations cannot be used to deduce the impacts of other missions
where the TriG receiver will be employed for science applications. Specific orbit and antenna
configuration for other TriG missions will need to be considered in order to make a holistic
determination of the maximum tolerance values (e.g., maximum transmitter power and total
number of LTE BS) for the terrestrial LTE network.
As an example of this, while the maximum antenna gain for Sentinel-6 is lower than the antenna
configuration used for COSMIC-2B, the percent time above the threshold (-73 dBm) is greater.
This is due to the higher orbit altitude of the Sentinel-6 providing a much larger field-of-view of
the U.S. resulting in a larger total number of LTE BS in view. This results in a lower peak
interference power due to greater space loss, but greater average power.
Also, consideration needs to be given to the reported results on TIME, when assessing the
overall impact of the scientific measurements. The typical duration for an atmospheric sounding
using RO is only one (1) to two (2) minutes. In certain modeled LTE BS deployment scenarios,
the average duration of an interference event may be well above the one (1) or two (2) minutes
needed perform an occultation measurement. Loss of any data during a given RO event will
generally result in loss of the entire RO profile.
Further, in conjunction with the TIME aspect, the time associated when a satellite in-orbit has
the continental U.S. (and adjacent areas in surrounding bodies of water – Atlantic Ocean, Pacific
Ocean, and Gulf of Mexico) within its field-of-view needs to be considered. As the reported
results under the % Time Threshold are referenced to a timeframe that is representative of a
satellite orbiting the entire Earth, the deduced interference time to occultations performed over
the continental U.S. will be significant. The results tables indicate that the TriG receiver will
effectively be degraded ( threshold) up to 9.6% of the time during a 10-day orbit. Since the
continental U.S. (and surrounding bodies of water) represent approximately 2.5% of the surface
of the earth, this represents a significant degradation to the ability for the TriG receiver in
providing valuable scientific data.
Finally, it is important to note that the results generated here use a 50% utilization factor. In a
time of weather crisis such as a hurricane, the utilization may increase substantially in response
to emergency responder and population communication needs. This increase in utilization would
increase the interference by 3dB at a time when the occultation measurements would be useful in
predicting a storm’s intensity and direction.
Impact Plots
It is important to provide a visual representation of the areas affected by the aggregate
interference received by the terrestrial LTE BS network. Figure K-18 through Figure K-20
depict the areas where the TriG receiver will be impacted for COSMIC-2B and Sentinel-6,
respectively. These figures are provided as a sample - additional impact plots for all of the
simulations run for COSMIC-2B, and for a majority of the simulations for Sentinel-6, are
available.

46
COSMIC-2B Impact Plots
Figure K-18 demonstrates the locations of where the TriG receiver will receiver various levels
of interference. In the COSMIC-2B plots, below, the colored levels are defined as:
Table K-16: COSMIC-2B Impact Plot Threshold Levels
Received Interference
Level (dBm)
Color Comment
-90 None Below simulation parameters. No interference recorded.
≤ -90 -73 YELLOW Interference received, but below -73 dBm (-1 dB C/No) threshold
≤ -73 -59 ORANGE
Interference received above -73 dBm (-1 dB C/No) threshold, but
below -66 dBm (loss-of-lock ) threshold
≥ -59 RED Interference received causes TriG to lose lock
Although the impact plots provide a representation of the areas where degraded performance of
the TriG receiver will occur, it must be noted that the position of the degradation signifies the
location of the LEO satellite and not where the occultation measurement is taking place, which is
about 3,200 km (COSMIC-2BB) or 4,200 km (Sentinel-6) before and behind the satellite.
The plots demonstrate the received interference levels based on the simulation parameters.
Therefore, the plots for COSMIC-2B depict simulations with the forward antenna subarray. RO
measurements from the forward array will slightly skew the overlaid interference plots toward
the equator, while the satellite is traversing in the southwest to northeast direction. In converse,
the RO measurements from the forward array while traversing over the U.S. in a northwest to
southeast direction will skew the overlaid interference plots toward the north pole.
Additional consideration must be given to the aft antenna array for COSMIC-2B, in combination
with these plots.

47
Figure K-18: COSMIC-2B Interference Impact Plot for Simulation 1a (Sample Plot)
Sentinel-6 Impact Plots
Figure K-19 and Figure K-20 demonstrate the interference impacts to Sentinel-6. Figure K-19
depicts the positions where the RO measurements are located when the interference power is
greater than 73 dBm. Coupled with Figure K-19, Figure K-20 depicts the level of received
interference from the LTE BS network as experienced at the satellite as it passes over the
Continental United States. These two (2) plots have been generated for each of the Sentinel-6
simulation runs performed with simulation 3c depicted.

48
Figure K-19: Sentinel-6 Plot of RO Locations for Simulation 3c (Sample Plot): Occultation
measurement locations affected when interference power is greater than 73 dBm

49
Figure K-20: Sentinel-6 Interference Impact Plot of Simulation 3c (Sample Plot) This plot
shows the level of interference in dBm experienced when the satellite passes over the
Continental United States.

DOT consists of multiple modal Operating

More Related Content

Similar to DOT consists of multiple modal Operating (20)

Recently uploaded (20)

DOT consists of multiple modal Operating