Initial tuning umts (2)

Limited Internal 2008-02-141
Emerson Eduardo Rodrigues
Eng Emerson Eduardo Rodrigues
01th December 2010

Initial Tuning is an integral part of the Radio Network
Design Process

Initial Tuning – Overview
The purpose of Initial Tuning is to ensure the radio network
performs according to the RND before entering commercial
service.
Initial tuning in general is performed on an unloaded
network or under limited load with friendly user traffic,
unless there is a specific request from the customer.

Initial Tuning
Customer IT Overview
Preparation & Radio
Network Audit
Drive Route
Definition
DriveTest
Post
Processing
Analysis
Change
Proposal
Site Testing
Update RF
Database
Reporting
Individual Site Drive
Cluster Drive
Cluster Drive Analysis

Initial Tuning - Preparation & Radio Audit
The preparation phase consists of the following activities:
- Definition of Clusters*
- Definition of RNC Boundaries*
- Definition of Drive Test Routes*
- Drive Test schedule & Resources
- Collect Radio Network Design information
- Setup test numbers and servers
- Equipment check
* Customer must be involved in these steps

Preparatory step: Network audit
 Prior to any action a network audit should be
performed
 A network audit typically include the following:
 Characterization of radio properties (pilots etc.)
 Complete neighbor relations overview
 Complete feeder check
 Complete parameter value & consistency check
 Counter statistics collection: starting point for
performance
 Setting up routine for monitoring all other activities in
the network (SW upgrades, reparenting etc.)

Initial Tuning - Site Testing
Performed prior to detail drive testing of any cluster, while
waiting for the cluster to achieve 90% completion, on a site
by site basis.
Main objectives:
• Ensure that the site is built according to the site design.
• Identify swapped feeders & antenna configuration issues
• Perform HO, cell change & reselection tests to identify
problems.
• Verify RSCP level.
• Verify CS and PS call establishment in all sectors.
• Basic Field Test and Functionalities for each service
(Speech, CS64, HS).

Initial Tuning - Preparation & Radio Audit
Activity Responsible Tool Time
Site Testing RF State TEMS Investigation 2 hrs per site
Tilt Check RF State Map Info and
Installation Report /
A955 3G
Alarm & Cell Availability Check + Iu/Iub
checks
UTRAN M2000 20 mins
Full Parameter Check Integration M2000 < 1 day
Radio-Only Parameter Consistency Check RF State A955, M2000 40 mins
Scrambling Code Collisions Check RF State A955 30 mins
Neighbour Cell Check RF State A955 3 hours
Handset Software Version Check RF State UE SW Tracker 5 mins
RSSI Check RF State M2000 10 mins
Traces Setup UTRAN M2000
The purpose of the Radio Audit is to find inconsistencies in
the network and fix them prior to drive testing, saving time,
rework and speeding up the tuning process:

Initial Tuning Drive Test
INITIAL TUNING
PREREQUISITES ACTIVITIES RESULTS
1. All planned sites
integrated, tested and
operational per cluster
(90% complete is a
minimum requirement)
2. Radio Network Design
implemented (including
PSC plan and Ncells)
3. Network is not in
commercial service
4. Clusters for verifying are
frozen
5. Single site checks
completed.
1. Preparation
2. Consistency Check
3. Drive test route plan
4. Data Collection
5. Post-processing
6. Analysis/Change request
7. Reporting
1. Verification that all
problems identified
during the analysis
phase (faults, coverage,
drop calls, missing
Ncells) have been
addressed.
2. Initial Tuning Analysis
report written including
identification of trouble
spots, RF related plots,
KPI stats and any
change requests issued
3. Presentation of results
The following table describes the activities involved in initial
tuning drive testing

Initial Tuning – DT Equipment
Aditional itens:
- Updated Cell File;
- Call Sequence File (Voice & Packet)
- Scanner Configuration;
- Maps.
USB
Scanner/GPS
Speech
Short
Video
Long PS GPS: Positioning
MS1 - Scanner : Measurements on frequency band
MS2 - Short Calls: Speech
MS3 - Long Calls: Video
MS4 - PS: Packet SwitchHUB / USB
USB

Initial Tuning Drive Test
Internal Process
* Estimated time for proposed changes implementation
Preparation Drive Test
Post
Processing
Analysis
Report
1 day 1-2 days 1 day 3 days 3 days*
1 day
Change
Proposal /
Verification
KPI
Acceptance?
Y
NY
N
Report
Accepted?
(customer)
Y
Cluster
OK
N
2 days
Update RF Database
(A955 and cell_Check)

Initial Tuning – Process Overview
Total (Max) = 24 ~ 28 days/cluster
ActivitiesTime plan
Drive 1
Scanner+ Speech (Short &
Long) + Video (Long) + PS
Drive 2
Cluster Acceptance Setup*
Drive 3 (Optional)
Drive (1-2 days)
Post Processing & Analysis (3 days)
Implement changes (3 days)
Drive (1-2 day)
Drive Analysis (1 day)
Drive (1-2 day)
Post Processing & Analysis (3 days)
Drive (1-2 day)
Drive Analysis (3 days)
Report PraparationReport (1 day)
* To be updated after Claro approval
Preparation (1 day)
Check Alarms, cell availability,
start UETR, etc…

Initial Tuning Flow

Initial Tuning
Analysis Process - Tools
TiPP Ver 4.8
CR Management

Initial Tuning 3G Environment
Analysis Workflow and Tools

2nd priority
Key Aspects to Consider
 Parameter consistency
 Consistent network
configuration
 Transport network capacity
 Cell planning, coverage
 Cell planning, interference
 Neighbor definitions
 Detecting crossed
feeders
 LA/RA planning
 Power settings
 Code plan
 L3 Analysis (Regional Team/ High Level)
 Feature tuning
1st priority
3rd priority

The Cell Search Procedure
Initiate Cell Synchronization
UE monitors Primary SCH code, detects peak in matched filter output
Slot Synchronization Determined ------>
UE monitors Secondary SCH code, detects SCG and frame start time offset
Frame Synchronization and Code Group Determined ------>
UE determines Scrambling Code by correlating all possible codes in group
Scrambling Code Determined ------>
UE monitors and decodes BCH data
BCH data, Super-frame synchronization determined ------>
P-CCPCH
(PSC + SSC + BCH)
UE adjusts transmit timing to match timing of BS + 1.5 Chips
Cell Synchronization complete

Cell planning - Coverage
Description
 Insufficient coverage can be classified as
 Incomplete coverage (coverage holes)
uplink or downlink
 Not possible to maintain/set up call
 Poor coverage
 Call maintained but below quality
threshold
 Coverage is dependent on load
 Load must always be considered
 Insufficient coverage (up- and/or
downlink) gives rise to
 dropped calls,
 reduced accessibility,
 inferior packet throughput
 low voice quality.
RAB
Required
Ec/No
unloaded
Required
Ec/No
With load
Required
Ec
unloaded
Required
Ec
With load
AMR 12.2 -16 -12 -116 -112
P64/64 -14 -10 -110 -107
P64/128 -12 -8 -110 -107
P64/384 -9 -5 -110 -107
A-DCH DL -16 -12 - -
A-DCH UL - -
16 kbps - - -117 -113
32 kbps - - -114 -110
48 kbps - - -112 -108
64 kbps -111 -107
128 kbps - - -108 -104
256 kbps - - -105 -101
384 kbps - - -103 -99
Example of typical thresholds for poor coverage – for
details see [4]
1st priority

Cell planning – Coverage
Ways to handle
 General tuning by:
 Antenna tilts
 Antenna redirections
 New Sites
 Short site to site distance will
increase the possibility of obtaining
coverage
above the defined levels.
1st priority

Cell planning – interference
Description
 Description: Interference can be classified as [1]
 “Pilot pollution” in idle mode
 More than one pilot at equal signal strenth
 “Pilot pollution” in active mode
 Excessive amount of pilots competing for the active set
 High interference
 more than 4 pilots within a window of 5 dB and satisfactory coverage
 Excessive interference results in
 Low system capacity
 Dropped calls
 Reduced accessibility
 Inferior packet throughput
 Low voice quality
1st priority

Cell planning – interference
Ways to handle
 Down-tilt/redirect antenna
 If necessary turn off interfering
sector
 Careful site placement
 Avoid high sites
 Careful antenna alignment
Reducing pilot power is not a
preferred solution to reduce
interference
1st priority
High sites generate interference
outside the planned coverage area

Neighbour definitions
Description
 Proper neighbour definitions are
essential
 A missing neighbour relation can
cause call drop
 Common reasons for poor non-
working neighbours
 Neighbour has not been defined
“missing neighbour”
 The neighbour list has been
truncated (too many defined
relations) best server
cell not in active set
Pilot Ec/No UE released
releaseConnOffset
1st priority
A1
C3
C2
C1
B3
B2
B1
G3
G2

Neighbour definitions
Ways to handle
 Defining neighbor relations
 Neighbor cell definitions must be consistent, including inter-
RNC relationships.
 Maximum 20 neighbor relations defined per cell, as a rule of
thumb
 Neighbor cell relation consistency checking must be performed
as a regular scheduled activity
 Re-parenting should be performed with OSS-RC to ensure
correct handling of inter-RNC neighbor relations, but relations
should be checked afterwards anyway.
 Strict rules for responsibility and approval of neighbor relations.
 Identifying missing neighbor relations
 Drive testing –
 Capture missing relations, only street level performance
 Statistics –
 Capture unused relations
 Neighbor cell support//TEMS Visualization –
 Identify missing AND unused relations
1st priority
A1
C3
C2
C1
B3
B2
B1
G3
G2

Parameter Consistency Keeping consistent parameter
setting is essential for a well
functioning network
 Non consistent setting is a
major source of quality and
performance issues
 Perform system parameter
dump every day
 Consistency check every week
 Special focus on parameters
that are set to different values
in different cells.
1st priority

Consistent network configuration
 Description:
 The software status consists of release, track, correction updates, emergency
upgrades, parameter scripts etc.
 Software status not under control may result in:
 Errors that have already been solved will reappear, errors that may be difficult to
trace.
 Dropped calls, reduced capacity and reduced accessibility
 Ways to handle
 The nodes must be loaded with the latest software,
 The software status must be known.
 The complete packages in the software releases must be used, i.e. including all
scripts aligning non configurable parameters etc.
 All RNCs should have the same software status.
 All WBTSs should have the same software status.
 Additional scripts to load parameters etc. must be known and under control
regarding if and when they were run.
1st priority

Transport Network Capacity
 Description:
 For HSDPA deployment transport capacity is often the
limitation
 May result in:
 Poor throughput performance
 Ways to handle:
 Ensure enough transport resources are configured
1st priority

Location, Routing and RNC area planning
Description
 Designing LA, RA and RNC areas that can handle
LA/RA updating, paging and Iur handovers
 Less careful area planning may result in:
 dropped calls
 paging congestion
 high RNC load
2nd priority

G3
Location, Routing and RNC area planning
Ways to handle
 Careful geographical placement of
LA/RA borders to minimize the number
of LA/RA updates and Iur handovers:
 in low traffic areas
 perpendicular to main traffic flow.
 Number of cells in LA/RA is trade-off
between
 load from LA/RA updates
 paging load
 Rules of thumb [5]:
 1 RNC per LA and RA initially
 same RA as LA
 NO sharing of LA/RA i.e. separate
LA/RA identities for GSM/GPRS and for
UMTS even if they cover the same
geographical area and the same sites in
case of co siting
A1
B3
B2
B1
C3
RNC 1
RNC 2
Place RNC border perpendicular to major
roads
2nd priority

WBTS
ASC/
TMA
ANT
ANT port
WBTS ant port
ASC/TMA ant port
WBTS
ANT
Config B
(with ASC/TMA)
ASC/TMA WBTS p
Config A
(no ASC/TMA)
B: Ref. Pt.
TX/RX power
A: Ref. Pt.
TX/RX power
A: Ref. Pt.
TX/RX power
Feeder att. = 0
OR
ANT feeder
ANT jumper
ANT feeder
Power Settings Inserting correct feederloss
parameter and correct power
setting at the system reference
point is important [2]
 To get desired coverage and
capacity
2nd priority
System reference point
- all power values are
referred at this point

A1
B3
B2
B1
G3
G2
A1
C3
C2
C1
B3
B2
B1
G3
G2
Monitored set
Code planning Proper code planning is important for well
functioning system
 Improper code planning may result in
 Code collision will generate interference and
dropped call
 Same code in monitored set may generate
drop call
 To avoid:
 Use all 512 codes
 Use consistency check rules to see that
 a cell and its neighbor cells do not have
the same code assigned
 a cell do not have two neighbor cells
using the same code
 Watch out for unusual propagation and
topographical conditions that may cause
distant cells to hear each other
Same code must
not appear in the
same monitored
set
13
13
5
5
Same code in two
adjacent cells will cause
high interference
2nd priority

Crossed feeder
 Description
 At installation approximately 10-20% of sites have
swapped feeders. RX and TX branch for different cells
may be mixed up.
 May result in:
 Poor UL performance, no coverage and strange behavior
 Ways to handle
 Scanner measurements of the pilot channels and/or test
UE measurements, and comparing to the network plan,
will catch around 80% of the cases
2nd priority

Detecting crossed feeder
Common scenarios
DL
UL
Reference case
DL
UL
No coverage, loss of UL diversity
Low signal level, poor Ec/No
Dropped calls
Handover fails
Call set up problems
High UL power
Swapped TX&RX
By monitoring Code, signal level, quality
and UE performance (handover, setup success,
power) crossed feeder can be detected.
RX2
TX/RX1
DL
UL
Swapped TX (or Swapped RX)
Loss of UL diversity
 High UL power
2nd priority

Case Symptom Monitoring tool
1 Scrambling codes cover wrong directions. NetAct
Handover may fail from other cells to them due to improper handover relationship or uplink DPCH synchronization
problem.
Connection setup will fail during random access or uplink DPCH synchronization procedures.
TEMS
2 No downlink coverage, i.e. low RSCP in some areas.
High downlink interference, i.e. low Ec/No and high DL RSSI in some areas.
Scrambling codes cover wrong directions.
NetAct
If the UE tries to connect to cell B in the cell A area, connection setup may fail during random access or uplink DPCH
synchronization procedures.
If the UE tries to handover to cell B in the cell A area, it may always send additional handover events to UTRAN but
handover function always fails due to uplink DPCH synchronization problems.
The UE connected to cell A transmits slightly higher UE Tx power than in the normal feeder case due to higher UL
interference, i.e. higher UL RSSI.
The connection may drop if the UE moves to the planned cell B area due to no coverage.
TEMS
3 The UE connected to cell A and/or cell B transmits slightly higher UE Tx power than in the normal feeder case due to
higher UL interference, i.e. higher UL RSSI.
TEMS
4 Connection setup will fail in both cells during random access or uplink DPCH synchronization procedures.
Handover will fail from other cells to either cell A or cell B due to uplink DPCH synchronization problems or improper
handover relationship.
TEMS
5 No downlink coverage, i.e. low RSCP in some areas
High downlink interference, i.e. low Ec/No and high RSSI in some areas
Scrambling codes cover wrong directions
NetAct
Connection setup will fail in cell A during random access or uplink DPCH synchronization procedures.
Connection may drop if the UE moves to the planned cell B area due to no coverage.
Handover will fail from other cells to either cell A due to uplink DPCH synchronization problem or improper handover
relationship.
The UE connected to cell B transmits slightly higher UE Tx power than in the normal feeder case due to higher UL
interference, i.e. higher UL RSSI.
TEMS

Detecting crossed feeder
Common scenarios
 Examples of monitoring tools for swapped feeder
problems are NetAct and TEMS. The symptoms can be
high downlink interference, slightly high UE Tx power,
connection setup failure during random access or
uplink DPCH synchronization procedures, no
downlink coverage, handover failure, wrong
scrambling code coverage, etc.

Parameter tuning
 Description:
 Field experiments to optimize the setting of selected parameters
 Systematic varying of parameters to evaluate “best” parameter setting.
 Expected effects:
 Greater understanding of functions and their behaviour
 Improved performance
 Ways to handle:
 The entities to handle should be functions rather than parameters
 What is a proposed parameter set to make this particular function to behave a certain
way?
 Then selecting a variety of “behaviors” to test
 Careful design of experiment and measurement principles
 Evaluate by counter statistics if viable considering traffic
3rd priority

Some Other Learnings
 Advantage of the lack of subscribers during pre-launch should be taken. Tuning the
network after launch will result in 20% more effort required to tune it.
 It is not recommended to launch the network with less than 90% of the planned sites up
and running in an area. Sites added after launch or the tuning activity implies that all
surrounding sites need to be retuned.
 Get Access to database on “Trouble reports”. This will allow the tuning team to correlate
issues revealed in the tuning activity with those found by the sysTEMS specialists in the
rollout of networks.
 The close cooperation of the Radio Design Consultants and Field Maintenance people will
play a vital role in resolving network instabilities or hardware problems seem in the network.
 Review the original drive routes for tuning, in some cases the drive time can be reduced
without degrading the test results.
 The type of parameter tuning is dependant on the load in the network.
 A detailed site audit should be held initially on all sites and all hardware. Issues identified
should be corrected prior to the tuning activity.

Some Other Learnings- Continued….
 Optimize the radio environment before tuning the parameters. The recommended parameters
hold well in most instances and the investigation of parameters on both an WBTS and RNC
level is best made once the RF environment has been addressed.
 Coordinate the hardware and software upgrades and patching schedule with the tuning
schedule. The impacts of network upgrades on the performance can severely degrade
the drive test results and in addition cause delays. In the worst case situation a
redrive will be required for all WBTS’s.
 Poor neighbor cell planning has resulted in a dropped call rate of 30% in the worst case. This
was found through the tuning activity and would have been disastrous for the launch
network performance.
 Generally the tuning activity has increased the performance of retainability and accessibility.
 Without tuning activities coverage holes that are not predicted in planning activities can
result in poor performance. In some cases coverage holes have been as prevalent as 15%.

Initial Tuning
Examples using Tools and Macros
1. TEMS Investigation
2. Route Analysis
3. Net Act-3g
4. TiPP-W
5. MAPINFO
6. GENEX/NASTAR
A955

Initial Tuning
Analysis Workflow & Tools
Define
Trouble Shooting
Areas
TiPP/MAPINFO, A9553g,
NetACT
Antenna
Config.
Optimization
A9553g, TiPP, MAPINFO,
NetACT or Antenna Tilt
Tool
NCell List
Optimization
TiPP, A955, Ncell_Check
or other Macros
UE Analysis
Dropped & Blocked
Analysis
Genex Inv. & RA, TiPP,
UETR, A955
Cluster Acceptance
Report;
CR
Recommendations
GENEX, NASTAR,
A955TiPP/MapInfo or
other Templates
END
Layer 3 Analysis
may be performed
by team in
Columbia/ High
Level analysis

Initial Tuning
Analysis Overview
1) Define Trouble Shooting Areas
2) Verify Pilot and Uplink coverage of cluster/cell (RSCP and
Ec/Io), identifying and analysing:
• Best Server Cells (Swapped Sectors) – TiPP/MapInfo
• Individual Cell SC Coverage to Identify Overshooting
Cells - TiPP/MapInfo
• Poor Coverage Areas (Poor RSCP and/or Poor Ec/Io) -
TiPP/MapInfo
• Pilot Pollution Areas - TiPP/MapInfo
• High UeTx pwr Areas - UE Analysis, TiPP
• High DL BLER Areas - UE Analysis, TiPP
• Low Throughput Areas - UE Analysis, TiPP

Initial Tuning
Analysis Overview
3) Define physical changes.
4) List High Detected Missing Neighbour Cells and propose
Neighbour List Changes.
5) Detailed dropped call and blocked/access analysis. Define
Neighbour List Priority
6) Identify abnormal system/UE problems (report to
Troubleshooting Team)
7) Produce “Cluster Acceptance Report” (and plots), fill up
the “KPIs Drive Test - Summarized Report” and update RF
database (A955/NetACT and Ncell) and propose physical
changes.

Initial Tuning
Propose Changes
Changes performed during the IT stage are limited to:
• Soft Parameters: Missing Neighbour Cells, Primary SC.
• Antenna system Configuration:
1) ET, identification of swapped feeders
2) MT, direction
3) Height, antenna changes, RF hardware such as filters and
amplifiers, etc… (not recommended at IT Phase).
• System related parameter changes: To be investigated and
implemented by a specialist team.
Abnormal behaviour caused by system HW/SW problems or UE are not within the scope of initial tuning.
These should be identified by regional teams and handled by a dedicated trouble shooting team.

Tuning Analysis

Pilot Coverage Levels:
Tuning Analysis
Define Trouble Shooting Areas
Pilot Coverage
Class
Conditions
Level 1 Ec/No >= -8 dB AND RSCP >= -87 dBm
Level 2
(Ec/No >= -12 dB AND RSCP >= -93 dBm) AND (Ec/No < -8 dB
OR RSCP < -87 dBm)
Level 3
(Ec/No >= -15 dB AND RSCP >= -107 dBm) AND (Ec/No < -12
dB OR RSCP < -93 dBm)
Level 4 Ec/No < -15 dB OR RSCP < -107 dBm

Pilot Coverage RSCP x Ec/No (Scanner)
Tuning Analysis
Pilot Coverage Analysis
UARFCN = 1062
RSCP
>=
-87
-87 >
RSCP
>= -93
-93 > RSCP
>= -107
RSCP <
-107
Any
RSCP
Coverage
Class
Percent
age
Ec/No >= -8 86.75 3.63 0.00 0.00 90.38 Level 1 86.75
-8 > Ec/No >= -12 6.15 1.42 0.00 0.00 7.57 Level 2 11.20
-12 > Ec/No >= -15 1.26 0.00 0.00 0.00 1.26 Level 3 1.26
Ec/No < -15 0.79 0.00 0.00 0.00 0.79 Level 4 0.79
Any Ec/No 94.95 5.05 0.00 0.00 100.0 Total 100.00

Pilot tuning – total RSCP for system
Tick the box for
best server
RSCP
seems
OK!
Analysing the logfiles with Map Info after Tipp *
*Pantalla solamente
para efectos
ilustrativos

Pilot tuning – total Ec/No for system
Tick the box for
best server
Ec/No,
small
problems
Analysing the logfiles with Map Info after Tipp*
*Pantalla solamente
para efectos
ilustrativos

Pilot tuning – Coverage per SC
Tick the
box for
best
server
Problems: Many
cells being the
best server in a
small area
Analysing the logfiles with Map Info after Tipp *
*Pantalla solamente
para efectos
ilustrativos

Pilot tuning – Active set size, zoomed view
Tick the
box
Problems:
Yellow dot = 5
cells in active
set size
*Pantalla solamente
para efectos
ilustrativos

Pilot tuning – Display interfering cells, 1
1: Click here
2: Remember to
choose which
drive-data to
look at
3: Click
here to get
the + cursor
Analysing the logfiles with Tems

Pilot tuning – Display interfering cells, 2
1: Click the
yellow dot
2: Cells covering
this dot are
listed
There should be a line
to each covering cell,
but that does not work
in this project. Cells are
marked with
instead.
Analysing the logfiles with Tems

Pilot tuning – Analysis of interfering cell 144
3: The cell 144 (blue)
seems to cover
backwards. Backlobe
or reflexions? Check
e.g. photos from Site
Survey. Redirect
antenna? Tilt?
1: Tick box to
enable ”Show
Coverage per SC”
view
2: In ”Show
Coverage per SC”
view, enter SC,
choose RSCP or
Ec/No plot.
4: Cell 144 (blue) also
has many neighbours
(green) backwards.
Redirect antenna? Tilt?

Pilot tuning – Checking antenna direction and tilt
1: Tick box to enable
”Site database” view
4: No mechanical or
electrical tilt! Site-to-site
distance is approx. 400
meters. The problem
area is in a valley.
Possible solution: Tilt to 8
degrees electrical.
2: Mark the
intended site
3: Mark
intended cell
But 144 was best server at
some points. What will happen
with RSCP and Ec/No after tilt?

Pilot tuning – Checking RSCP and Ec/No after tilt
1: Tick box.
Remember
to look at
the 2nd best
server.
3: Possible solution:
Tilt the sectors pointing to
the problem area to
improve Ec/No.
Remember that 144 will
be tilted, also improving
Ec/No.
2: RSCP OK,
but size of
square shows
Ec/No not OK.
(It can be
better to show
RSCP and
Ec/No
separately.)

Tuning Analysis : MAPINFO PLOT using TiPP Output files
Define Trouble Shooting Areas
Pilot Coverage Levels:

Tuning Analysis
Best Server and SC Verification
Best Server and SC Verification can give an indication of
the following problems:
• Swapped TX feeders.
• Cells not transmitting or with transmission issues.
• No Best Serving cells defined areas.
• Overshooting cells.

The SC plan and overshooting issues are preferably
analyzed in TiPP/MapInfo & TEMS Tools.
TEMS
Log files
BR DUMPRNC Dump
Pilot pollution
SC coverage
Best Server plot
Report to
Implementation
Swapped Feeder
or Wrong SC
Analyze coverage
per SC
End
Change
Proposal
No (Report Cells)Yes
No
Are cells
transmitting?
SC
appears
next to its
cell?
TiPP/MapInfo
SC as
planned?
TEMS
Route
Analysis
Overshooting
Sector?
Yes
No
Yes
Yes
No
Yes
Tuning Analysis
Best Server and PSC Plan Verification

Tuning Analysis
Best Server and SC Verification
Individual
Plot
Analyze:
• Swapped TX feeders.
• Cells not transmitting or with
transmission issues.
• No Best Serving cells defined areas.
Plot from Melon Tool

Tuning Analysis
Individual SC 08 Plot Example

Tuning Analysis
Overshooting Sector (Individual SC 17 Plot Example)

Tuning Analysis
Pilot Coverage: Area where the CPICH from the best serving
cell is detectable, i.e. received by an UE with sufficient quality
and signal strength to be able to camp on the cell. Pilot
coverage holes are areas where this condition is not met.
Pilot channel RSCP or Ec/Io is also used during cell
selection/reselection, and for handover decisions.
MapInfo/TEMS RA & TiPP/MapInfo tools are used (Scanner or
UE) to detect and locate pilot coverage holes over the drive-
test route.

Pilot Coverage RSCP x Ec/No (Scanner)
Tuning Analysis
UARFCN = 1062
RSCP
>=
-87
-87 >
RSCP
>= -93
-93 > RSCP
>= -107
RSCP <
-107
Any
RSCP
Coverage
Class
Percenta
ge
Ec/No >= -8 86.75 3.63 0.00 0.00 90.38 Level 1 86.75
-8 > Ec/No >= -12 6.15 1.42 0.00 0.00 7.57 Level 2 11.20
-12 > Ec/No >= -15 1.26 0.00 0.00 0.00 1.26 Level 3 1.26
Ec/No < -15 0.79 0.00 0.00 0.00 0.79 Level 4 0.79
Any Ec/No 94.95 5.05 0.00 0.00 100.0 Total 100.00

Tuning Analysis
Pilot Coverage RSCP_Ec/No x Problem Areas (Scanner)

Tuning Analysis

Tuning Analysis
Pilot Coverage Analysis TiPP Plot
Pilot Coverage Ec x Ec/Io (Scanner)

Tuning Analysis
Pilot Coverage Analysis TiPP Plot
Pilot Coverage RSCP (Scanner)

Tuning Analysis
Pilot Coverage Analysis (Other Market Example)

Plots UE TX Pwr
Tuning Analysis
Pilot Coverage Analysis (Other Market Example)

The Effect of Optimization on Improving RF Environment
Tuning Analysis

Pilot pollution: Detection of many high power pilots as
compared to Best Serving Pilot that do not contribute
to the received signal.
The UE has the ability to constructively use signals in
soft/softer handover, all the other signals received that
exceeds the Active Set act as interfering. This
interference degrades the performance of the system.
Tuning Analysis
Pilot Pollution Definition

Reason: No Dominant Cell
Solutions:
- Remove cells overlapping by tilts/pans or reducing
CPICH
- Increase CPICH of desired cell
Tuning Analysis
Pilot Pollution

Reason: Dominant Interferer
Solutions:
-Remove cells overlapping by tilts/pans or reducing
CPICH
- Adding the overshooting cell to the neighbor list.
- Increase CPICH of desired cell
Tuning Analysis
Pilot Pollution

–100
–90
–80
–70
–60
–50
–40
–30
–25 –20 –15 –10 –5 0
Ec/N0 [dB]
RSCP[dBm]
Interference limited Both design criteria
fulfilled
Coverage limitedCoverage and interference
limited
Areas with high interference can be detected by correlating
low EcNo with high RSCP.
Possible solutions:
• Down-tilt/redirect antenna
• If necessary turn off interfering sector
• Reducing pilot power is not a preferred solution
Tuning Analysis
Pilot Pollution

Tuning Analysis
Pilot Pollution x Problem Areas – Scanner Plots
 Displays areas with 4 or more pilots within a set dB (“Pilot Pollution
Threshold”) of the strongest cell
 Measured by the NetAct
 Severity Level represents the number of pollutants

Tuning Analysis
Pilot Pollution - Overshooting Sector may cause Pilot Pollution

Prior to evaluating site changes, it is a good practice to
review:
• Antenna change restrictions depending on co-location and
construction. Many cells will only be allowed ET, and some
have azimuth change restrictions due to cluster mount kits.
• Site photos (panorama)
• The height of the antenna/building/surrounding buildings,
obstructions or risk for shadowing (not allowing excessive
down tilt)
• The antenna type, pattern and current settings of
mechanical/electrical tilt.
• Analyze coverage using A955 3G or other Planning Tool
Tuning Analysis
Solutions to Coverage and Interference Problems

Tuning Analysis

Tuning Analysis
Attention when suggesting tilt
changes:
A 3° downtilt can lead to
different results if the tilt is
changed from 0° to 3° or from
5° to 8° (p.e.), depending on
the antenna pattern.
>15dB
2 dB

System reference
point:
- all values are
refeered in this
point.
Tuning Analysis
System Reference Point for Power Settings

Changing CPICH power to solve coverage problems
may have the following drawbacks and potential
disadvantages:
• Due to uneven pilot power settings, some UEs might no
longer be connected to the “closest” cell in terms of path
loss. This will cause near–far problems and
consequently increase the uplink interference.
• Since DPCH powers are assigned relative to the pilot
channel, an increase in pilot power will increase the
DPCH power and may cause blocking.
Tuning Analysis
Solutions to Coverage and Interference Problems (Power
Settings)

Changing CPICH power to solve coverage problems
may have the following drawbacks and potential
disadvantages (cont.):
• The cell with increased pilot power will absorb more UEs
from adjacent cells, thus increasing its own load.
• Finally, increasing pilot power may cause uplink
coverage and pilot coverage imbalances. See picture on
next slide.
Tuning Analysis
Solutions to Coverage and Interference Problems (Power
Settings)

Tuning Analysis
UL & Pilot Coverage Imbalance
Increasing pilot power may cause uplink coverage and pilot
coverage imbalances

Tuning Analysis
Missing Neighbour Analysis
best server
cell not in active set
Pilot Ec/No UE released
releaseConnOffset [12 dB]
 Good neighbour definitions are essential
 Missing neighbour relation can cause call drops
Common reasons for poor non-working neighbours
 Neighbour has not been defined=> “missing neighbour
 The neighbour list has been truncated

Tuning Analysis
Running Missing Neighbor Analysis using the Post Processing
Tool (TiPP/MapInfo) will show which neighbors are missing most
frequently and which should be reprioritized.
Once cells are identified for review, further analysis usually with
help of TiPP/MapInfo should focus on whether it is feasible to add
this neighbor to the neighboring cell list.

Neighbour Prioritisation
 Uses scanner and cel file to tally the number of “hits”: times a handover
will occur between each neighbour relation
 Takes existing cel file as basis and colours depending on relation state
 White = existing relation used
 Yellow = existing relation not used (may not be required)
 Red = missing neighbour needed (consider adding)
 Can use to reprioritise neighbours to avoid truncation effect
Tuning Analysis

Missing Neighbour Analysis (1)
 Based on exact A955 UMTS 3G Missing Neighbour events detected by
the scanner
 Scanner uses cel file to determine if an undefined NCELL is stronger
than defined ones

Missing Neighbours Analysis
 Takes Raw MN results and groups missing relations in order of “hits”
 Many filters set in “Input Index” sheet used to remove less important
MN relations
 When serving cell has poor Ec, EcIo
 If missing relation is within a certain dB of the strongest
 If missing relation occurs a certain number of times
 Distance between 2 sites also provided

Detected Neighbour Analysis
 Another form of missing neighbour detection (by UE)
 Detected Set (DN): cells detected by UE, but are included neither in
Active Set nor Monitored Set
 Good indicator of missing neighbours or truncation effect (hence neighbour
reprioritisation required)
 Default settings: strong pilot = better -12, weak pilot = between -12 and -
18.
 Priority 1 = Weak AS and good DN
 Priority 2 = Good AS and good DN

Detected Neighbour Analysis
 Takes Raw DN results from all UEs and groups detected relations in
order of priority, then number of “hits”
 Distance between 2 sites also provided

Active set (AS)
max 32 cells
Cell A
Cell B
Cell C
Cell A1
Cell B1
Cell C1
Cell A10
Cell B10
Cell X
Cell Y
Cell Z
Monitored set
Unmonitored set – not measured by all UE
Neighbours are truncated
Avoiding truncation is important to keep drop rate low
A healthy network has not more than 15-20 relations per cell
Tuning Analysis
Truncated Neighbour

Tuning Analysis
SC Clashes
 If same SC from 2 different cells appear in neighbour
list (1st, 2nd, 3rd tier)
 Use c3 Tool or Mapinfo to find
macro
inbldg
inbldg
macro
neighbour
neighbour
neighbour
ActiveSetSC = 14 SC = 14

Tuning Analysis
NL Database

UE Analysis

UE Analysis
Methods
The following method is recommended to help analyze
calls:
• Identify problems. Use the Post Processing Tools
visualized events. They can be overlaid on Pilot
Pollution or Coverage plots.
• Analyze in Tems Investigation. This can be done
by replaying the log files in Tems Investigation.
- Use Signaling Flow Diagrams to find out where a call has
gone wrong.
- Synchronize A955 and TEMS to track the signaling in both
directions and see where the failure occurred.

UE Analysis
Methods (cont.)
• Classify the problems. Classify the UE dropped and
blocked calls into the following broad categories:
- RF Related
* Missing 3G Neighbour;
* Missing IRAT Neighbour;
* Pilot Pollution;
* Poor Coverage;
* Fast Fading/rapidly changing RF.
- Site Installation Issues / Sites Not Integrated
- Hardware and Software Problems
- UE Issues
- System Related Issues (e.g. Core, Utran, E1, Admission, Alarms)
- Tems and A955 Investigation / User Issues
- Unknown

• Propose Solutions. A physical change, neighbour
addition or parameter change may be required to fix
the problem. Alternately if a solution cannot be
found it should be tracked.
• Track Problems. A master spread sheet should be
kept (on the network drive), mainly for use by the
National team to follow up with further detailed
analysis of abnormal drops and blocks.
UE Analysis
Methods (cont.)

UE Analysis
Plots Ec_Ec/Io Scanner x Call Drop

UE Analysis
Accessibility and Retainability
Accessibility and Retainability are two important KPIs that
can be measured from UE data.
Accessibility performance can be measured by call setup
success rate. Call setup success rate is counted as the
number of successful call set-ups divided by the total
number of call setup attempts.
Performance Indicator Typical Target Value
CS Speech Call Setup Success Rate > 97%
CS Video Call Setup Success Rate > 97%
PS Successful RAB Establishment > 97%

Common Reasons for Call Setup Failure:
- Cell Locked (Check cell availability);
- Cell Blocked (Check admission thresholds, feeder loss, DL tx
power);
- Max UE TX power reached (UL coverage / HW problem);
- DL code power max, then DL coverage or HW problem;
- Initial DL SIR target set too low;
- RF Issues (High RSSI, pilot pollution, low RSCP, etc…)
- Missing Neighbour Issues;
- SW or parameter issues (Ex: High RAB setup failures)
Check L3 messages to identify when the failure occurred.
UE Analysis

TEMS L3 messages can identify when the failure occurred
(Examples):
- Random Access Failures
- RRC Connection Complete not sent
- Call Proceeding not received
- Alert or Connect not received
- RRC Connection Setup not received
- Measurement Control not received
- Radio Bearer Setup not received
Identifying where, when and the number of failures occurred
(on a sector basis) will help on troubleshooting.
UE Analysis

Retainability is defined as the ability of the user to
establish and maintain a service (voice, video or pkt call).
Performance Indicator Typical Target Value
CS Speech Drop Call Rate < 2%
CS Video Drop Call Rate < 2%
Pkt flow session drop (R99 or
HSDPA)
<2%
UE Analysis

Common Reasons for Dropped call:
- Cells Locked or Blocked (Check admission Thresholds, feeder loss, DL
tx power and cell availability, alarms, etc…);
- DL code power max (DL coverage or HW problem);
- High DL Transport Channel BLER;
- Max UE TX power reached (UL coverage / HW / Interference problems);
- RF Issues (High RSSI, pilot pollution, low RSCP, etc…)
- Missing & Truncated Neighbour Issues;
Check TEMS L3 messages to identify when the failure occurred.
Identifying where, when and the number of failures occurred (on a sector
basis) will help on troubleshooting.
UE Analysis

 TiPP Tool will generate some a sheet that can be used to locate the
dropped and blocked call events.
UE Analysis
Failed Events

Integrity
Analysis

Integrity
Performance Indicator
Typical Target
Value
CS Call Setup Time 95th Percentile TBD
CS Speech Quality DL BLER <= 2%
CS Video Quality DL BLER <= 2%
Hard Inter-RAT UMTSGSM Handover
Success Rate
> 97%
Hard Inter-RAT GSMUMTS Handover
Success Rate
> 97%
The integrity of a service can be defined by its performance
once the network has been accessed. The table below lists
some integrity metrics along with their target values:

Block Error Rate – TiPP-W
 Records BLER messages (every 2 sec) on dedicated transport channel
depending on call type
 BLER samples binned any size and average/percentile calculated
 BLER samples before a dropped call can be removed by setting a timer
with the “Filter seconds before drop” threshold
 Separate for each UE
 Also mappable to see areas of poor quality

Handover Analysis

UMTS RAN HO:
 Soft/Softer Handover
• Soft Handover: UE connection consists of at least two radio links
established with cells belonging to different WBTS.
• Softer Handover: UE connection consists of at least two radio links
established with cells belonging to the same WBTS.
 Inter-Frequency Handover
• UE is moving out of coverage of one UMTS RAN frequency to an
area where coverage of another UMTS RAN frequency exists.
 Inter Radio Access Technology (Inter-RAT) handover
• UE is on dedicated channels for circuit switched services.
• UE is moving out from UMTS RAN coverage to an area where only
GSM/GPRS network coverage exists.
 Inter Radio Access Technology (Inter-RAT) cell change
• UE is on dedicated channels for packet switched services.
• UE is moving out from UMTS RAN coverage to an area where
only GSM/GPRS network coverage exists.

Signaling flow when changing the
Active Set
RNC UE
Perform
Measurement
RNC Evaluation
UE Evaluation
Radio Link Addition
MEASUREMENT CONTROL message
(DCCH)
MEASUREMENT REPORT message
(DCCH)
Radio Link
Add/Remove/Replace
Radio Link Removal
RNC Evaluation
MEASUREMENT CONTROL message
(DCCH)
ACTIVE SET UPDATE
(DCCH)
ACTIVE SET
UPDATE COMPLETE
Execution

UE Measurement Concepts & Features
 Event-Triggered Reporting
UMTS Handover is based on event-triggered reporting
• UE is in CELL_DCH state
 Event 1a, A Primary CPICH enters the Reporting Range
 P-CPICH, not included in the Active Set,
• measured P-CPICH Ec/No > (P-CPICH Ec/No of the Best Cell in
the Active Set) - reportingRange1a + hysteresis1a/2
• At least during a time equal to timeToTrigger1a
• Only one cell with highest Ec/No is considered and retained
• If the Present Cells in Active Set < maxActiveSet => the cell is
proposed to be added to the Active Set
• If the Active Set is full => the cell is proposed as a replacement of the
worst cell (the reported cell should has a better quality than the worst cell in the
Active Set)

 Event 1b, A Primary CPICH leaves the Reporting Range
 P-CPICH, included in the Active Set,
• measured P-CPICH Ec/No < (P-CPICH Ec/No of the Best Cell
in the Active Set) - reportingRange1b + hysteresis1b/2
• At least during a time equal to timeToTrigger1b
• If the report includes more than one cell, handover algorithms will
remove the reported cells one by one from the Active Set, however
one cell is always kept in the Active Set for maintaining the
connection.

Reporting event 1a and 1b (Add and
delete)
reportingRange1a
Measurement
quantity
time
P_CPICH best cell
reportingRange1b
P_CPICH 2
C_TTT1a C_TTT1b
Hysteresis1a/2
Hysteresis1a/2
Hysteresis1b/2
Hysteresis1b/2
6* (6X0.5) = 3dB
10* (10X0.5) = 5dB
0*(0X0.5)
= 0dB
C_TTT1b = 0.2* secC_TTT1a = 0.2* sec
Reporting event
1b
Reporting event
1a
* Current default values

 Event 1c, A non-active primary CPICH
becomes better than an Active Primary CPICH
 P-CPICH, not included in the Active Set,
• measured P-CPICH Ec/No > (P-CPICH
Ec/No of the Weakest Cell in the Active
Set) + hysteresis1c/2
• At least during a time equal to
timeToTrigger1c
 If the report contains more than one cell
fulfilling 1c criteria, only the one with highest
Ec/No is considered and retained. If the
retained cell is a valid cell, and the Active Set is
full, the cell is proposed as a replacement for
the weakest cell in the Active Set. .

Reporting of event 1c (replace)
hyst1c/2
Measurement
quantity
time
P_CPICH 1
Reporting event 1c
P_CPICH 2
P_CPICH 3
P_CPICH 4
C_TTT1c
hyst1c/2
Hyst 1c = 2* (2X0.5) = 1dB
0.5*dB
0.5*dB
C_TTT1c = 0.2* sec

 Event 1d, Change of Best Cell
 P-CPICH, included in the Active Set,
Monitored Set or Detected Set, or P-CPICH
is not included in the Active Set
• measured P-CPICH Ec/No > (P-
CPICH Ec/No of the Best Cell in the
Active Set) + hysteresis1d/2
• At least during a time equal to
timeToTrigger1d
• If P_CPICH is already in the current Active
Set, a MEASUREMENT REPORT message
is sent to inform that the best cell in the
Active Set has changed
• If P_CPICH did not belong previously to
the current Active Set, it would be proposed
to change the worst cell in the Active Set

Reporting of event 1d (change of
best cell)Measurement quantity
time
P_CPICH 1
Reporting event 1d
P_CPICH 2
P_CPICH 3
C_TTT1d
Hysteresis1d/2
Hysteresis1d/2
C_TTT1c = 2.560* sec
Periodical reporting if
replace fails
Hyst 1d = 2* (2X0.5) = 1dB

Handover Analysis
The objective of handover analysis is to make sure that
successful handover is observed mutually between defined
neighbours.

 Initial design in planning phase
 Use site data and planning tool
 Based on power order of received CPICH
 Modification of neighbour list during initial tuning
 Use scanner data from cluster drive
 Use post-processing tool such as TEMS RA and
TiPP/MapInfo to find missing neighbors
 Should be investigated together with tilt optimization as
it can change significantly. First set tilts, then drive to
determine neighbours
 Optimize neighbour list after launch
 Based on traffic data
Handover Analysis
Neighbour List Process

Handover Analysis
Neighbour List Checks
 May need to add/delete neighbours based on local
knowledge
 Add cells on same site, close by even if planning tool
does not indicate
 Delete cells from far away sites, should downtilt them
instead as it will cause interference after the active set is
full.
 Don’t forget to add symmetrically

Neighbor Set (intra-frequency)
Active Set
Unmonitored Set (intra-frequency)
Detected Set (intra-frequency)
cells actively used in a service, in soft/softer handover
the union of the neighboring cells of the cells in the AS,
excluding the cell in AS
cells in the Neighbor Set that the UE is not ordered
explicitely to measure on
Intra-frequency cells detected by the UE but not part
of Active Set or monitored set
Monitored Set (intra-frequency)
cells measured by the UE, but not part of the AS.
Handover Analysis
Neighbour Relation

Handover Analysis
Creating Monitored Set (1)
 Active Set cells added in EcNo order
 RNC then takes the highest priority neighbours of each active set
member, until 32 cells in monitored set
 For each step, if the neighbour SC has already been added, then
ignore it and move to the next step
 All instances of active set cells in other lists will be ignored
 All duplicates in neighbour lists will be ignored removed
 When the 32-list is full, the remaining neighbours that are negated

Handover Analysis
Creating Monitored Set (2)

 Upper limit on maximum number of intra frequency
cells that can be measured by a UE is 32, hence 31
neighbours
 only a certain number of neighbors from each active
set cell are included in the monitored subset
 UE can also measure cells in detected set to find
strong cells that are not defined as neighbors or
excluded in the monitored subset reduction
 Recommend 15-20 neighbours per cell in order of
priority
Handover Analysis
Sorting Neighbour Lists

Handover Analysis
Intra/Inter RNC HO Failure
Handover failures may be due of one of the following:
- Poor uplink quality (Link Unbalanced?);
- Incorrect handover series (the desired cell is not defined in
the neighbor list of the unwanted cell);
- Missing neighbor relationship ;
- Improper cell individual offset settings;
- Small soft handover area;
- High active set update rate (for several possible reasons):
* Improper handover parameter settings.
* Pilot pollution in dedicated mode
* Overlapping Pilot Coverage
- Truncated Neighbour List.
- SC clashes

Clutter Statistics: Cells_Within_Margin (5dB)
<=1 <=2 <=3 <=4 <=5
before 31.85% 62.54% 83.54% 93.53% 97.99%
after 41.30% 72.63% 89.39% 96.39% 98.73%
improved 9.45% 10.09% 5.85% 2.86% 0.74%
Handover Analysis
Active Set Size Performance Monitoring

Measurement Quantity
(Ec/No and RSSI)
UMTS cell
GSM cell
usedFreqTresh2fEcno
usedFreqTresh2dEcno
hysteresis2d/2
hysteresis2f/2
hysteresis2d/2
utranTresh3aEcno
gsmTresh3a
hysteresis3a/2
Reporting
event 2d
Reporting
event 2d
Reporting
event 2f
Reporting
event 3a
=> Start of
Compressed Mode
=> Stop Compressed
Mode
Reporting event 2d,2f and 3a (interRAT)
HO to GSM
ttt3a
Both conditions satisfied:
GSMthresh3a,
utranthresh3a
The event 3a is triggered when the estimated quality of the UMTS RAN used
frequency is below the threshold usedFreqThresh2dRscp+utranRelThresh3aRscp -
hysteresis3a/2
and the measured GSM carrier RSSI of a GSM/GPRS cell is above the absolute
threshold gsmThresh3a, during a time at least equal to TimeToTrigger3a;
Handover Analysis IRAT

SRNC CN/UMSC Target BSS
CELL_DCH
1. DCCH (AM) ”Measurement Report” (Event 2d)
2. Evaluation of MR
3. Compressed Mode Control
4. DCCH (AM) ”Measurement Control”
5. DCCH (AM) ”Measurement Report” (Event 3a)
6. Evaluation of MR
7. ”Relocation Required”
8. GSM ”Handover Request”
9. GSM ”Handover Request Ack”
10. ”Relocation Command”
11. DCCH (AM) ”Handover from UTRAN Command”
GSM HO access + HO complete
12. GSM ”Handover Complete”
13. Iu Connection Rlease
14. Dedicated radio link release
A successful IRATHO from UMTS to
GSM

 Symptom: I-RAT HO and cell change functions
activated, but no HO when EcIo is low
 Reason: Improper parameter settings
 event 2d and/or UMTS RAN threshold too close to –16
 need to set to higher value
 if too high, more users in compressed mode
 Event 3a too high
 set to lower value
 If too low, more congestion on GSM
Handover Analysis
IRAT Failure (example)

Event 2d

Switch to Compressed Mode

Measurement Control with GSM information

Measurement Report e3a

RRC Handover from
UTRAN Command

Tuning Analysis
Tools

NetACT C3/NeXplorer TEMS RA TiPP/MapInfo UETR ACTIX?
Design Overview 
Parameter Consistency 
RSCP / EcNo 
Intereference / Pilot Polution  
Mobility 
Co Scrambling Codes
Missing Neighbour  
Accessibility/ Retainability 
Throughput  
UL BLER, SIR 
Tuning Analysis
Tools

Important parameters to Tune
during Initial Tuning
moc_name node_id param_name
AntennaBranch WBTS mechanicalAntennaTilt
RetDevice WBTS electricalAntennaTilt
Sector WBTS beamDirection
Sector WBTS height
SectorAntenna WBTS antennaType
CoverageRelation RNC utranCellRef
CoverageRelation RNC hsPathlossThreshold
CoverageRelation RNC coverageIndicator
GsmRelation RNC externalGsmCellRef
GsmRelation RNC selectionPriority
Handover RNC maxActiveSet
UtranCell RNC primaryScramblingCode
UtranCell RNC maximumTransmissionPower
UtranCell RNC qHyst1
UtranCell RNC qHyst2
UtranCell RNC hoType
UtranRelation RNC utranCellRef
UtranRelation RNC selectionPriority

L3 Signalling
Channels and Protocols

L3 Signalling
Channels
 Logical Channels –
Defines what
information to be
transferred
 Transport Channels –
Defines how data
should be transferred.
 Physical Channels –
Carry the data

L2/MAC
L1
L2/RLC
L3/RRC
UTRANUE
RRC
RLC
MAC
PHY
RRC
MAC
PHY
RLC RLC RLC
Signaling
Radio Bearer
Radio Bearer
Logical Channel
Transport Channel
Physical Channel
CTRL CTRLUSER
DATA
USER
DATA
L3 Signalling
Channels

L3 Signalling
Protocol
- The Control Signaling is used for troubleshooting the problems. These
messages are carried in the Logical Control Channels such as DCCH,
CCCH, BCCH.
- Radio Resource Control (RRC) Protocol messages contain the necessary
control information and the commands – Obtained from TEMS.
- In the Iub we may be interested in the NBAP and RANAP protocol
messages – Obtained from UETR.
PHY
RLC/MAC
RRC
MM GMM
CC SM
PHY
RLC/MAC
RRC
Sig. Bearer
SCCP
RANAP
UE WRAN
Sig. Bearer
SCCP
RANAP
MM GMM
CC SM
MSC/VLR or SGSN

BASIC Call Flow Sequence and Algorithms

L3 Signalling
Intra RNC Soft Handover
Remember !!! – Unlike GSM, the measurement reports
event triggered. You can see long times of silence from
the UE. These are not to be confused as missing
messages.
UE RNC
RRC: Measurement Report (DCCH)
RRC: Measurement Control (BCCH/DCCH)
UE Measures the SCs
and performs
Evaluation for Event
Trigger Criteria.
RRC: Active Set Update (DCCH)
RNC Evaluates based
on the measurements
and Network
Configuration.
RRC: Active Set Update Complete (DCCH)

 Cause codes will indicate the problem if RRC Connection
is rejected
UE RNC
RRC: RRC Connection Request (CCCH)
RRC: RRC Connection Setup (CCCH)
RRC: RRC Connection Setup Complete (DCCH)
Cause Code for connection
Request Sent:
•Originating Streaming Call
• Originating Interactive Call
• Originating Background Call
• Terminating Conversational Call
• Terminating Streaming Call
• Terminating Interactive Call
• Terminating Background Call
• Emergency Call
• Registration
• Originating High Priority
Signalling
• Call re-establishment
• Terminating High Priority
Signalling
• Terminating cause unknown
Contains the UE Capability (HS or
R99)
RRC Connection can
be rejected to
Admission &
Congestion Control
Features
L3 Signalling
RRC Connection Establishment

T
>T300
UE gets RRC
Connection Setup
UE sends RRC Connection
Setup Complete
N = N + 1
T=0 N > N300
Fail. UE goes into idle
mode
UE gets RRC
Connection Reject
Cause = “Wait‟
Wait for „wait‟ seconds
N = N + 1
T=0
UE sends RRC
Connection Request
N
N
N
Y
Y
Y
Y Y
N
N
RRC Procedure
Successful
L3 Signalling
RRC Connection Establishment
UE sends RRC
Connection Request
UE increments Timer T
UE starts timer T=0,
counter N=1

L3 Signalling
Radio Bearer Setup
 RAB (Radio Access Bearer) is setup between Core and UE.
 Radio Bearer is setup between UE and UTRAN to support
the RAB.
UE RNC
RRC: Radio Bearer Setup (DCCH)
RRC: Radio Bearer Setup Complete (DCCH)
CN
RANAP: Radio Access
Bearer Request
RANAP: Radio Access Bearer
Response (DCCH)

 UTRAN sends Active Set Update to the UE in CELL_DCH
mode.
 The result could be one of the following
 Radio Link Addition
 Radio Link Removal
 Radio Link Replacement (Combined Addition/Removal)
UE RNC
RRC: Measurement Report (e1a, e1b, e1c) (DCCH)
Timeout of 5 s
L3 Signalling
Active Set Update

L3 Signalling
Radio Bearer Reconfiguration
 RB Reconfiguration occurs for the following scenarios
 Channel Type Switch i.e., DCH to FACH, FACH to DCH
 Reconfiguration of the Radio Bearer after RB Setup (such
as RLC timer reconfiguration)
UE RNC
RRC: Radio Bearer Reconfiguration (DCCH)
RRC: Radio Bearer Reconfiguration Complete (DCCH)
FACH/DCH
DCH/FACH

L3 Signalling
Transport Channel Reconfiguration
 Transport Channel Reconfiguration procedure
triggers the Channel Rate Switching. 64K <-> 128K <-
> 384K
UE RNC
RRC: Transport Channel Reconfiguration (DCCH)
RRC: Transport Channel Reconfiguration Complete
(DCCH)

L3 Signalling
RRC Connection Release
 When the RRC Connection is released, the UE will be in
IDLE mode. Radio Bearers will be released.
 RRC Connection can be released abnormally
UE RNC
RRC: RRC Connection Release (DCCH / CCCH)
RRC: RRC Connection Release Complete (CCCH)
RRC: RRC Connection Abnormal Release (DCCH /
CCCH)

(MOC)
UE RNCRRC: RRC Connection Request (CCCH)
RRC: Initial Direct Transfer - CM Service Request (DCCH)
RRC: Downlink Direct Transfer – Auth & Ciph Request (DCCH)
RRC: Uplink Direct Transfer - Auth & Ciph Response (DCCH)
RRC: Security Mode Command (DCCH)
RRC: Security Mode Complete (DCCH)
RRC: Uplink Direct Transfer - SETUP (DCCH)
RRC: Downlink Direct Transfer – CALL PROCEEDING (DCCH)
RRC: Downlink Direct Transfer – ALERTING (DCCH)
RRC: Downlink Direct Transfer - CONNECT (DCCH)
RRC: Downlink Direct Transfer – CONNECT ACK (DCCH)
RRC: Measurement Control (DCCH)
Call Setup Phase
KPIs:
•Call Setup
Success Rate
•Call Setup Time
RRC: Uplink Direct Transfer - DISCONNECT (DCCH)
RRC: Direct Transfer - Release (DCCH)
RRC: Uplink Direct Transfer – Release Complete (DCCH)
Total Call
Time
KPIs:
•Dropped
Call
RRC
Connection is
released after
the Call
Release.

Example PS R99 Call Flow
RRC: Initial Direct Transfer - CM Service Request (DCCH)
RRC: Downlink Direct Transfer – Auth & Ciph Request (DCCH)
RRC: Uplink Direct Transfer - Auth & Ciph Response (DCCH)
RRC: Security Mode Command (DCCH)
RRC: Security Mode Complete (DCCH)
RRC: Uplink Direct Transfer – SM ACTIVATE PDP CONTEXT
REQUEST (DCCH)
RRC: Downlink Direct Transfer – SM ACTIVATE PDP CONTEXT
ACCEPT (DCCH)
RRC: Measurement Control (DCCH)
RRC: Uplink Direct Transfer – SM Deactivate PDP Context Request
(DCCH)
RRC: Direct Transfer – SM Deactivate PDP Context Accept (DCCH)
RRC: RRC Connection Release (DCCH)
Call Setup Phase
KPIs:
•Call Setup
Success Rate
•Call Setup Time
Total Call Time
KPIs:
•Dropped Call
RRC: Transport Channel Reconfiguration (DCCH)
RRC: Transport Channel Reconfiguration Complete (DCCH)
Channel Switching

Example HSDPA Call Flow
RRC: Uplink Direct Transfer – SM ACTIVATE PDP CONTEXT
REQUEST (DCCH)
RRC: Downlink Direct Transfer – SM ACTIVATE PDP CONTEXT
ACCEPT (DCCH)
RRC: Measurement Report (e1a) (DCCH)
RRC: Physical Channel Reconfiguration (DCCH)
RRC: Physical Channel Reconfiguration Complete (DCCH)
E1A:
Radio Link
Addition
Not a HS Cell
Change. Here
the procedure is
to realign the
parameters.
RRC: Measurement Report (e1d) (DCCH)
RRC: Physical Channel Reconfiguration (DCCH)
RRC: Physical Channel Reconfiguration Complete (DCCH)
E1D:
Change of Best
Server
HS Cell ChangeKPI
•HS Cell Change
Success Rate
•HS Cell Change
time
KPI
•Call Setup
Success Rate
•Call Setup Time

Appendix A
RAN Tuning Report

Initial tuning umts (2)

More Related Content

What's hot (20)

Similar to Initial tuning umts (2) (20)

More from Emerson Eduardo Rodrigues Von Staffen (8)

Recently uploaded (20)

Initial tuning umts (2)