5G Fundamentals

5G Fundamentals
Damien Magoni
University of Bordeaux
2017/04/19
Version 3
1

Attribution
• The material contained inside is intended for teaching.
• This document is licensed under the CC BY-NC-SA license.
• Relevant sources are listed on the following References slide.
• All figures and text borrowed from these sources retain the rights of
their respective owners.
2/79

References
• Next Generation Mobile Networks (NGMN) Alliance, NGMN 5G White
Paper, v1.0 (17-02-2015)
• Recommendation ITU-R M.2083-0, IMT Vision – Framework and
overall objectives of the future development of IMT for 2020 and
beyond (09/2015)
• Signal Processing for 5G – Algorithms and Implementations, Fa-Long
Luo & Charlie Zhang (Eds), IEEE Press – Wiley, ISBN 9781119116479
(2016)
• What Will 5G Be?, Andrews et al., IEEE Journal on Selected Areas in
Communications, Special Issue on 5G Wireless Communication
Systems (14-05-2014)
3/79

Table of Contents
1. 5G definition, usages, requirements, architecture and timeline
2. Millimeter-wave wireless communications and channel frequencies
3. Massive MIMO and beamforming
4. In-band full duplex transmission and self interference cancellation
5. Ultra dense networks and Device-to-Device (D2D) communications
4/79

5G definition, usage scenarios,
requirements, architecture and
timeline
Part 1

Objectives
• 5G description and proposed visions
• 5G usage scenarios, business models, value creation and
requirements
• Necessary improvements and technology trends
• 5G overall architectural paradigm and timeline
6/79

5G description
• 5G stands for the fifth generation of mobile technology which should
address both consumer and business needs from 2020 and beyond.
• 5G is expected to enable a fully mobile and connected society and to
empower socio-economic transformations in countless ways many of
which are unimagined today, including those for productivity,
sustainability and well-being.
• 5G will operate in a highly heterogeneous environment, where there
is a fundamental need for 5G to achieve seamless and consistent user
experience across time and space.
7/79

Propositions for a 5G Vision
• NGMN 5G: the Next Generation Mobile Networks (NGMN) for 5G is a
white paper issued by the NGMN alliance in 2015.
• IMT-2020: the International Mobile Telecommunications for 2020 is a
recommendation issued by the Radiocommunications committee of
the International Telecommunication Union (ITU-R) in 2015.
8/79

Use Case Scenarios
NGMN 5G IMT-2020
5G will support countless emerging use cases with a high variety of applications and variability of their performance attributes.
9/79

Business Models
• Three roles are envisioned, each
one having several possible
business models
• Asset Providers can offer XaaS and
real-time network sharing.
• Connectivity Providers can offer best
effort and QoS differentiated feature
sets.
• Service Providers can either offer
integrated services enriched by
partners’ content to customers or can
allow partners to directly make offers
to customers enriched by the
operator network.
10/79

Value Creation
• 5G will provide services tailored
to each customer profile (i.e.,
consumers, enterprises, verticals
and other partnerships).
• 5G will provide value-added
enablers structured around
trust, experience and service
over a reliable and flexible
network.
11/79

Requirements Dimensions
• A 5G system should deliver a
consistent user experience (defined
by service-dependent minimum KPIs)
over time for a given service
everywhere the service is offered.
• The 5G use cases demand very diverse
and sometimes extreme requirements
grouped in six items shown in the
figure.
• A single solution to satisfy all the
extreme requirements at the same
time may lead to over-specification
and high cost. Thus, combinations of
multiple solutions may be more
effective.
12/79

Use Case Categories
• From the 8 scenarios, NGMN
derives 14 use case categories
shown in the table.
• Several use cases will be active
concurrently in the same
operator network, thus requiring
a high degree of flexibility and
scalability of the 5G network.
13/79

User Experience Requirements
For each use case category, a set of requirement values is given, which is representative of the
extreme use cases in the category. The importance of capabilities vs usage scenarios is also shown.
IMT-2020NGMN 5G
14/79

Necessary Improvements over 4G
NGMN 5G IMT-2020
The network capabilities of the baseline 4G system (3GPP Release-12) fall short of
the NGMN requirements in a number of areas.
15/79

Technology Trends
• Current trends, shown in the
figure, can help improve existing
4G systems. However, incremental
evolution of 4G systems will not be
sufficient to address all the
capabilities’ shortfalls.
• Thus, 5G systems will have to
introduce new breakthrough
technologies such as
• Millimeter waves modulation
• Massive MIMO
• Full duplex transmission
• Network densification
16/79

5G Architecture
• The 5G architecture is a native
SDN/NFV architecture covering
aspects ranging from devices,
infrastructure, network functions,
value enabling capabilities and all
the management functions to
orchestrate the 5G system.
• APIs are provided on the relevant
reference points to support
multiple use cases, value creation
and business models as illustrated
in the figure.
17/79

Network Slicing
• A network slice, namely “5G slice”,
supports the communication service of a
particular connection type with a specific
way of handling the Control-plane and
User-plane for this service.
• To this end, a 5G slice is composed of a
collection of 5G network functions and
specific radio access technologies (RAT)
settings that are combined together for
the specific use case or business model.
• The intention of a 5G slice is to provide
only the traffic treatment that is
necessary for the use case, and avoid all
other unnecessary functionality.
18/79

IMT-2020 Timeline
• The IMT-2020 vision has been articulated from 2014 to 2016.
• 5G standards are to be written from 2017 to 2020.
• 5G systems deployment will start in 2020 at the earliest.
19/79

Millimeter-wave wireless
communications, new waveforms
and allocated spectrum
Part 2

Objectives
• Millimeter-wave features
• Sensitivity to the geometry of the propagation environment
• Non-stationarity in time and space
• New waveforms
• Allocated frequencies
21/79

Millimeter Waves
• Terrestrial wireless communication systems
have largely restricted their operation to the
relatively slim range of microwave
frequencies that extends from several
hundred MHz to a few GHz corresponding to
wavelengths of a few centimeters up to 1
meter. This spectral band has become nearly
fully occupied, in particular at peak times and
in peak markets.
• Vast amounts of relatively idle spectrum do
exist in the “millimeter-wave” range of 30–
300 GHz with wavelengths of 1 to 10 mm.
• Those bands have the following unique
features
• high path loss
• sensitivity to propagation environments
• vulnerability to geometry blockage
• non-stationarity in time and space
22/79

High Path Loss
• Given Frii’s equation of radio propagation
in free space, an increase in signal
frequency by 10 times results in a
decrease in received power by 100 times.
• The absorption due to air and rain is
noticeable, especially the 15 dB/km
oxygen absorption within the 60-GHz
band (which is why this band is
unlicensed)
• It is inconsequential for the urban cellular
deployments currently envisioned where
base station spacings might be on the
order of 200 m. In fact, such absorption is
beneficial since it further attenuates
background interference from more
distant base stations, effectively
increasing the isolation of each cell.
23/79

Sensitivity to Propagation Environments
• Because of the short wavelength, small
objects in the propagation environment
almost invisible at lower frequency bands,
become prominent.
• Specular reflection can be undermined
due to diffuse scattering. New paths can
be created from reflections from small
objects.
• As the radio frequency increases, the
propagation behaves more like optical
propagation. The low diffraction
probability of quasi-optical propagation
leads to a high probability of blockage.
• Receivers at locations behind a building or
around a corner can be severely
attenuated.
24/79

Vulnerability to Geometry Blockage
• Calculated mm-wave base stations (BS)
associations with real building locations.
The shaded regions correspond to
association with the BS centered at that
shade.
• Blocking, line-of-sight (LOS) vs non-LOS
propagation, and beam directionality
render our usual notion of cell boundaries
obsolete (disk-shaped cells).
• Because of the sensitivity to blockages, a
given link can rapidly transition from
usable to unusable and, unlike small-scale
fading, large-scale obstructions cannot be
circumvented with standard small-scale
diversity countermeasures.
25/79

Non-stationarity in Time and Space
• In practical communication scenarios, the environment is dynamic:
people and cars are moving around. Small moving objects, mostly
invisible in the low frequency bands, will cause turbulence in high-
frequency-band propagation.
• The channel in the high frequency band becomes non-stationary.
Some of the paths can be temporarily blocked, while new paths could
be created due to reflection from passing objects.
• The statistics of the transient blockage and the new paths depend on
the traffic in the communication scenario.
26/79

Millimeter-wave challenges
• New channel models are needed for 5G as current models are not accurate
for mm-waves.
• Directional beamforming both at BS and user equipment (UE) is essential
to compensate for a large amount of path loss and, accordingly, to create a
reliable radio link.
• Therefore, the mm-wave 5G RAT is required to accommodate narrow
beam-based system operation from an initial access to the data
transmission.
• Propagation losses due to pathloss, blocking and air and rain absorption for
mm-wave frequencies are surmountable, but require large antenna arrays
to steer the beam energy and collect it coherently.
27/79

New waveforms for 5G
• Orthogonal frequency-division multiplexing (OFDM) is the access
scheme used in today’s LTE/LTE-Advanced networks but it is limited
by a high sensitivity to frequency and clock offsets.
• New waveforms are being considered as candidates for 5G
• generalized frequency-division multiplexing (GFDM)
• filter bank multicarrier (FBMC)
• universal filtered multicarrier (UFMC)
• filtered OFDM (f-OFDM)
28/79

Filter Bank Multi-Carrier (FBMC)
• OFDM suffers from inter carrier interference
(ICI) due to the use of rectangular filters that
create lobes in the frequency domain.
• A filter bank is an array of filters, which are
applied to synthesize multicarrier signals at
the transmitter and analyze received signals
at the receiver.
• The normalized time and frequency responses
of a filter bank show that it can be designed
to achieve a side-lobe as small as possible.
• In the frequency domain, the frequency
response of the FBMC subcarrier is very
compact. The ripples can be neglected, and
there is no ICI between the non-neighboring
subcarriers.
29/79

New Multiple Access Schemes
• New multiple access schemes are being
researched, including sparse code multiple access
(SCMA), non-orthogonal multiple access (NOMA),
and resource spread multiple access (RSMA).
• NOMA for example, introduces power-domain user
multiplexing, exploits channel-gain difference
among users to improve spectrum efficiency, and
relies on more advanced receivers for multi-user
signal separation at the receiver side.
• Users far from the BS with low signal-to-noise
ratios (SNR) can receive higher power signals
(upper blocks in the figure) while users close to the
BS with high SNR can receive smaller power signals
(lower blocks)
• Non-orthogonal user multiplexing using
superposition coding at the transmitter and
successive interference cancellation for getting the
correct blocks at the receiver not only outperforms
orthogonal multiplexing, but also is optimal in
achieving the capacity of the region.
30/79

ITU-R Regions
• The International Telecommunication
Union (ITU) divides the world into three
radio regulatory regions for the purpose
of managing and harmonizing global radio
spectrum allocation and use.
• Each region has its own set of frequency
allocations, which are generally very
similar.
• Region 1 comprises Europe, Africa, the
former Soviet Union, Mongolia, and the
Middle East west of the Persian Gulf,
including Iraq.
• Region 2 covers the Americas including
Greenland, and some of the eastern Pacific
Islands.
• Region 3 contains most of non-FSU Asia
east of and including Iran, and most of
Oceania.
31/79

ITU-R Allocation to 5G at WRC-19
• The World Radiocommunication Conference in 2019 (WRC-19) will consider the following bands for 5G
under the agenda item 1.13: 24.25-27.5 GHz, 31.8-33.4 GHz, 37-43.5 GHz, 45.5-50.2 GHz, 50.4-52.6 GHz, 66-
76 GHz, and 81-86 GHz.
• Some countries are investigating other mobile bands above 6 GHz not being considered at WRC-19. The 28
GHz band has been permitted for 5G use in the United States and is being examined by Japan and South
Korea. This would complement the 24 GHz band supported in the European Union.
• Spectrum within the 3.3-3.8 GHz range is regarded as the basis for initial commercial 5G services, and some
countries are exploring whether other bands could be used such as 3.8-4.2 GHz and 4-5 GHz.
• Many mobile bands in the 1-6 GHz range currently used for 3G/4G services could be re-farmed for 5G use.
32/79

FCC Allocated 4 Bands to 5G Mobile Services
in July 2016
• The 28 GHz licensed band currently assigned for LMDS (fixed broadband wireless service) and
FSS (Fixed Satellite Services) will be usable for mobile services in dense population centers
and along highway corridors.
• Most of the 37 GHz licensed band for terrestrial mobile operations will be available on a co-
equal basis to federal and non-federal uses.
• The 39 GHz licensed band is authorized for terrestrial mobile operations with co-allocation
for fixed satellite and mobile services. Service rules have been adopted to minimize the risk
of interference between terrestrial and satellite uses.
• The 64-71 GHz unlicensed band will serve as a place to test and deploy wireless devices,
contributing to the development of new unlicensed applications.
33/79

Massive MIMO Systems,
Beamforming
Part 3

Objectives
• Understand what are MIMO systems
• Define massive MIMO systems
• Describe linear precoding and detection schemes
• Understand beamforming
• Describe the use of beamforming for mm-waves
35/79

MIMO Systems
• Modern communication systems rely upon multiple antennas at the
transmitter and/or receiver to enhance link performance.
• This class of techniques, known as multiple input, multiple output (MIMO),
exploits the spatial dimension by employing spatial encoding and/or
decoding.
• Multiple antennas in these systems can be used to increase link robustness
by using space-time block codes or data rate by applying spatial-
multiplexing.
• MIMO techniques can be extended beyond point-to-point to multi-user
applications with multi-user MIMO (MU-MIMO). MU-MIMO can be used to
separate users by their spatial position, allowing for further network
densification and increased capacity.
36/79

Massive MIMO
• Massive MIMO are systems
where the number of antennas
at the base station is increased
by an order of magnitude or
more over current MIMO
systems.
• A base station is using M
antennas to spatially multiplex K
single-antenna terminals (K<<M)
• H is the channel state
information
37/79

Requirements
• The base station should have a good knowledge of the propagation
channel in both directions, on which efficient downlink precoders and
uplink detectors can be based.
• Since acquisition of channel-state information (CSI) is generally infeasible in
the downlink, massive MIMO systems typically rely on channel reciprocity,
uplink channel estimation, and time-division duplex (TDD).
• With the massive number of channels to estimate between base station
and mobile stations, a long-enough channel coherence time is needed to
allow for efficient operation. The accuracy at which we can estimate the
channel and the time interval over which it can be assumed constant bring
fundamental limitations to massive MIMO.
38/79

Baseband Processing for Massive MIMO
• The processing performed by an
massive MIMO base station that uses
OFDM signaling is shown on figure.
• M antennas of synchronized uplink
baseband samples are acquired by an
ADC and processed using an OFDM
receiver and then passed to a MIMO
detector and channel estimator.
• Channel estimates are used to
precode downlink data. Precoded
symbols are then distributed to the M
OFDM transmitters and transmitted
out the antenna ports.
39/79

Linear Precoding Schemes
• The downlink transmitted signals should be chosen such that users receive
their own symbols, with suppressed interference caused by the symbols
intended for other users.
• Given the massive MIMO assumption of an excessive number of base
station antennas, we can assume that linear precoding methods will work
well in scenarios where we have favorable propagation conditions.
• Current precoders F are linear precoding matrices calculated from the
channel estimate H supposed true and include
• the maximum-ratio transmission (MRT) precoder: F = α HH
• and the zero-forcing (ZF) precoder: F = β HH(H HH)−1
• where α and β are scalar parameters and HH is the conjugate transpose of H
40/79

Linear Detection Schemes
• The uplink signal model is easily derived by following a similar argumentation as
for the downlink, given that we assume the propagation channel to be reciprocal.
• A notable difference is that we do not perform any precoding on the user side,
since users are assumed not to cooperate to reduce interference. The only thing
the users can control is their own transmitted power level.
• Proper combining of signals from the M antennas can amplify desired signals and
reject interfering ones. Since the down- and uplink transmissions in TDD systems
take place over the same reciprocal channels, the same rates are typically
achievable in both directions.
• Precoding schemes therefore have counterparts called combiners in the uplink
detection, defined as a G matrix combining received signals from all antennas
• maximum-ratio combiner (MRC): G ∝ HH
• zero-forcing (ZF) combiner: G ∝ HH(H HH)−1
41/79

Massive MIMO Challenges
• The above description of precoding and detection techniques assumed that the
channel estimator delivered the true channel which is not correct. We need to
estimate the channel before we can precode or detect data.
• With N parallel narrowband systems (N-carrier OFDM), NMK channel coefficients
must be calculated. Given the non-stationarity of the channel updated 100 times
per s, hundreds of millions of channel coefficients per second need to be
calculated.
• Having a N-carrier OFDM system, using an L sample cyclic prefix, allows us to
reduce the dimensionality from N to roughly L per transmit/receive antenna pair,
in the worst case, across all subcarriers. This results in LMK parameters to
estimate, which can then be interpolated to the NMK channel coefficients.
• Given this dimensionality of the channel, we need to excite the channel with a
minimum of L pilots from each of the K terminals. These LK pilots are received by
all M base station antennas, giving the minimal number of LMK samples needed
for channel estimation.
42/79

Beamforming
• Beamforming is a traffic-signaling system for cellular base
stations that identifies the most efficient data-delivery route
to a particular user, and it reduces interference for nearby
users in the process.
• The primary challenge for massive MIMO is to reduce
interference while transmitting more information to many
users from many more antennas at once.
• Placing antennas in a 2-dimensional grid is called 3D-MIMO
and is an effective way to reduce the antenna panel size. It
can adjust the direction of the transmit beam in both the
horizontal and vertical dimensions, enhancing the spatial
resolution in the third dimension, so improving the signal
power and reducing inter-cell interference.
• With beamforming, stations can send individual data packets
in many different directions, bouncing them off buildings and
other objects in a precisely coordinated pattern and allowing
many users and antennas on a massive MIMO array to
exchange much more information at once.
43/79

Beamforming for mm-waves
• For millimeter waves, cellular signals are easily blocked by objects and tend
to weaken over long distances. Beamforming can help by focusing a signal
in a concentrated beam that points only in the direction of a user, rather
than broadcasting in many directions at once.
• This approach can strengthen the signal’s chances of arriving intact and
reduce interference for everyone else.
• Besides boosting data rates by broadcasting over millimeter waves and
beefing up spectrum efficiency with massive MIMO, wireless engineers
achieve the high throughput and low latency required for 5G through a
technology called full duplex, which modifies the way antennas deliver and
receive data.
44/79

Full Duplex Transmission, Self
Interference Cancellation
Part 4

Objectives
• Understand half and full duplex transmission
• Define self interference
• Understand self interference cancellation (SIC)
• Describe SIC techniques for SISO systems
• Present SIC challenges for MIMO systems
46/79

Half-Duplex Transmission
• Previous generations of mobile
communication mainly depended
on half-duplex transmission
schemes, in which the transmitted
and received signals are separated
either:
• in the time domain: time division
duplexing (TDD), as in Figure (a)
• in the frequency domain: frequency
division duplexing (FDD), as in Figure
(b)
• in both: half-duplex FDD as in Figure
(c).
47/79

In-band Full Duplex Transmission
• The term “full-duplex” (FD) was traditionally used when the device
had simultaneous bidirectional communication, in contrast to “half-
duplex” (HD), which assumed time-division duplexing.
• Previously, use of the term full-duplex assumed utilizing a pair of
frequencies to transmit and receive simultaneously.
• However, in recent years the term has carried a new concept: the
device can transmit and receive at the same time and over the same
frequency, as shown in Figure (d).
• Many papers use the term “in-band full duplex” (IBFD) to clarify this
new concept but most refer to it by an abbreviated version: “full
duplex”
48/79

Full Duplex Benefits
• Full-duplex can double the spectral efficiency (bit/second/Hz).
• Full-duplex can improve the reliability and flexibility of dynamic
spectrum allocation in wireless systems, such as cognitive radio
networks, either with in-band full duplex or partial band-overlap FDD
systems.
• Full-duplex can enable the small cells in 5G to reuse radio resources
simultaneously for access and backhaul.
49/79

Self-Interference Signal
• As shown in Figure (d), the major
challenge to the implementation of
full duplex is the self-interference
(SI) signal: the part of transmitted
signal that leaks into the receiver
chain.
• With the desired signal from the
remote node being weak, high-
power SI is a major issue for the
receiver.
• To achieve the best performance of
a full-duplex system, the SI signal
has to be suppressed to reach the
receiver’s noise floor.
50/79

Self-Interference Cancellation (SIC)
Techniques for SISO Systems
• passive SI suppression in the propagation
domain: conditional placement,
directivity, polarization and shielding
• active SIC in the analog domain: tapping
the transmitted analog signal and feeding
it with a negative sign to the receiver
• active SIC in the digital domain:
replication of the transmission samples
and feeding it with a negative sign to the
receiver
• auxiliary chain SIC: a hybrid technique of
the two previous methods, where
replication and cancellation domains are
different – one is analog and the other is
digital.
51/79

Passive SIC in the Propagation Domain
• Antenna separation is a method in which full-duplex systems
use different antennas for transmitting and receiving. The SIC
performance of antenna separation is limited by the physical
size of the transceiver.
• A circulator is a passive 3-port device in which an RF signal
entering any port is only transmitted to the next port in
rotation. For example, the signal that enters Port 1 is
transmitted to Port 2, but is isolated to Port 3, as in Figure (b).
Using this property, a circulator can be adopted as an isolator
for a single-antenna full-duplex system, which uses an
antenna for both the transmitter and receiver. When the
signal is transmitted through Port 1 to Port 2 where the
antenna is connected, it is not transmitted to the receiver at
Port 3.
• Antenna cancellation uses destructive interference to cancel
the self-interference. When a signal and the π-phase rotated
one is added, they cancel out each other. A π-phase shifter is
employed on one of the transmit antennas, which are
symmetrically placed at a distance d from the receive
antenna as in Figure (c).
• Isolation by dual-polarized antennas is to transmit and
receive in orthogonal polarization states thus avoiding
interference.
52/79

Antenna Cancellation
• Antenna cancellation uses destructive interference to cancel
the self-interference.
• When a signal and the π-phase rotated one is added, they
cancel out each other. By placing them λ/2 apart as shown in
the figure, the phase of the signal is rotated by π.
• In this method, however, the received powers of the two
signals are different due to asymmetric path loss.
• As a solution, a π-phase shifter is employed on one of the
transmit antennas, which are symmetrically placed at a
distance d from the receive antenna as in Figure (c).
• After passing through the same distance, the two transmitted
signals are assumed to be attenuated to the same degree.
Furthermore, antenna cancellation is applied not only using
two transmit antennas, but also two receive antennas.
• Combining two antenna cancellations in the transmitter and
receiver, additional isolation can be obtained.
53/79

Directional Isolation
• By suppressing the line-of-sight
component, these methods,
compared to others, can provide a
large amount of isolation.
• Directional isolation uses the
directional antennas, which are
arranged so that the gain of the
transmit antenna is low in the
direction of the receive antenna,
and vice versa.
• Such an approach would not work
for point-to-point full-duplex
scenarios
54/79

Antenna Shielding
• Electromagnetic shielding with copper
or aluminum plates can enhance the
isolation between antennas.
• However, one disadvantage is that the
shielding affects the far-field coverage
patterns because it prevents the
antenna from transmitting
to/receiving from the shielding
direction.
• Absorptive shielding (b) is preferred
on the reflective shielding (a) plates,
as the latter would couple with the
transmit antenna and subsequently
cause another component of self-
interference.
55/79

Orthogonal Polarization
• The polarization of an antenna is the
polarization of the radiated
electromagnetic fields produced by an
antenna, evaluated in the far field. For
perfect transmission, antennas must
be similarly polarized (a).
• if a horizontally polarized antenna is
trying to communicate with a
vertically polarized antenna, there will
be no reception (b).
• Isolation by dual-polarized antennas
(b) consists in transmitting and
receiving in orthogonal polarization
states thus avoiding interference.
56/79

Passive SIC Issues
• Although passive SI suppression techniques are appealing for reasons
of their simplicity, they are highly sensitive to the wireless
environment and its reflected paths, which cannot be known during
the design.
• Moreover, their effectiveness is greatly limited by the device form-
factor: the smaller the device, the less room there is to implement
such techniques
57/79

Active SIC Techniques
• Active SIC techniques adopt the following methodology: a replica of
the transmission signal is created and then adjusted to match the SI
channel, making the replica as similar to the SI signal as much as
possible, in order to subtract it from the total received signal.
• This copy can be created either in the analog domain or in the digital
domain before the digital-to-analog converter (DAC).
• The SIC signal stays in the same domain from where it was copied,
thus no additional ADC/DAC is required.
58/79

Active SIC in the Analog Domain
• Replication of the transmission signal in the analog
domain can be achieved by tapping the Tx chain,
using a power splitter, or using a balun (balanced–
unbalanced) circuit in the case of two separate
antennas as shown in the Figure.
• After creating an exact negative replication of the
signal (RF reference signal) from the inverter, the
replica is adjusted by delay and attenuation
elements to match the self-interference.
• Adjusting the signal is often achieved using a noise-
cancellation active chip. The chip takes the input
signal from the balun circuit, and separates it into
in-phase and quadrature components. A fixed
delay is applied to the quadrature component;
meanwhile any variable delay can be achieved by
controlling the gains of in-phase and quadrature
components. Adding this adjusted reference signal
to the total received signal from the Rx antenna
will partially cancel out the SI signal.
59/79

Active SIC in the Digital Domain
• Digital cancellation aims to cancel the residual
SI by applying advanced DSP algorithms to
process the SI signal and cancel both linear
and nonlinear components.
• Firstly the receiver decodes the unwanted
packet, reconstructs it and then subtracts it
from the originally received collided signal.
• For SI, a correlation operation is performed
without the need for decoding because the
unwanted packet – its own transmitted
samples – is already known to the receiver.
• The correlation between the received signal
and the clean transmitted signal is needed to
detect the peaks that give the path delay of
the SI channel.
60/79

Auxiliary Chain SIC
• (a) auxiliary Tx chain:
• It copies the baseband IQ samples of the transmitted
signal in the digital domain, then it uses an additional
transmitter chain to generate the SIC signal and feed it
back into the receiver in order to be subtracted from
the total received signal.
• The SIC signal has to be adjusted (pre-distorted) in
the digital domain before the DAC in order to
match the transmitted signal through the SI
channel.
• (b) auxiliary Rx chain:
• The transmitted signal is tapped in the analog domain
just before the antenna, and then fed back to the
receiver digital domain. However, like the auxiliary
transmitter method, the SIC signal is also adjusted in
the digital domain to exploit DSP algorithms. The
auxiliary receiver chain method mitigates the effects on
SIC of transmitter hardware impairments such as phase
noise and nonlinearities. A common oscillator for the
chains – ordinary and auxiliary – is used to mitigate the
phase-noise effect on the SIC signal.
61/79

Full-Duplex MIMO Systems
• Full-duplex MIMO may use one shared
antenna for each Tx–Rx pair with a
circulator, as shown in figure (a), in order
to strengthen its viability.
• However, when multiple circulators are in
place, severe interference among the
multiple shared antennas would occur.
• Multiple separate antennas have the
advantage of exploiting the degree of
freedom in the spatial domain as shown
in figure (b).
• So far, practical full-duplex MIMO system
designs, able to transmit and receive from
all antennas simultaneously with
sufficient SIC, are yet to be realized.
62/79

Ultra Dense Networks,
Device-to-Device (D2D)
Communications
Part 5

Objectives
• Network densification principle and small cells
• Interference management by network coordination and interference
cancellation receivers
• Mobility management by dual connectivity, virtual cells, virtual layers,
mobility anchors and handover command diversity
• Architecture and backhaul
64/79

Network Densification Principle
• To develop 5G, efforts have been made in
three areas:
• spectrum expansion
• spectrum efficiency
• and network densification.
• Network densification is considered to be
the paramount and dominant approach
to address the data challenge.
• It can be achieved by deploying a large
number of small cells – microcells,
picocells, femtocells, relay nodes and WiFi
access points – which are low-powered
radio access nodes and have smaller
coverage areas than macrocells.
65/79

Heterogeneous Small Cells
• An effective way to increase the network capacity is to make the cells
smaller. Cell shrinking enables the reuse of spectrum across a geographic
area and the ensuing reduction in the number of users competing for
resources at each base station (BS).
• As networks will become increasingly heterogeneous, a key feature will be
increased integration between different RATs, with a typical 5G-enabled
device having radios capable of also supporting 3G and 4G, several types of
WiFi, and perhaps direct device-to-device (D2D) communication, all across
many spectral bands.
• Hence, determining which standard(s) and spectrum to utilize and which
Base Station(s) or users to associate with will be a truly complex task for
the network.
66/79

Interference Management
• Network Coordination (network side)
• Spatial-domain coordination: coordinated multi-point techniques (CoMP) and interference
alignment techniques, use of arrays of transmit antennas or active antennas with coordinated
beamforming and joint transmission between cells
• Time-domain coordination: cells are time-synchronized and coordinate at which time-instances
they transmit, such that there are time-instances where Cell A can serve its users without
interference from Cell B. This is also known as coordinated muting.
• Frequency-domain coordination: hard or soft frequency reuse between neighboring cells. The
frequency-domain resource partitioning can be on physical resource block resolution, or on carrier
resolution if there are networks with multiple carriers.
• Power-domain coordination: adjustment of transmit power per cell to improve the interference
conditions.
• Advanced receivers (user side)
• Linear receiver with interference suppression: by means of linear combining of received signals at
the users’ antennas.
• Non-linear receiver with interference cancellation: the user estimates one or multiple interfering
signals and subtracts them from the received signal, followed by detection of the desired signal
67/79

Coordinated Multi-Point Transmission and
Reception
• CoMP transmission/reception is an effective method to
improve both average and cell-edge throughput by mitigating
the Inter Cell Interference.
• Categorized by the different number of points for
coordinated transmission, CoMP techniques include:
• single-point coordinated transmission by coordinated
scheduling/beamforming (CS/CB)
• multiple-point joint processing (JP)
• In CS/CB, by single-point transmission, coordination of
scheduling decisions and transmit-beam selection among
multiple cells is used to reduce inter-cell interference (ICI).
• In JP, data are shared among multiple points for joint
transmission to one or more users simultaneously so as to
achieve a performance gain by signal combining and
interference nulling from multiple points.
• For JP, there are two schemes: joint transmission (JT) and
dynamic point selection (DPS) and dynamic point blanking
(DPB).
68/79

Interference Alignment
• The fundamental concept of interference
alignment (IA) is to align the interference
signals in a particular subspace at each
receiver so that an interference-free
orthogonal subspace can be solely
allocated for data transmission.
• This is a promising technique to efficiently
mitigate interference and to enhance the
capacity of UDNs.
• In an L-cell system with one user in each
cell, assume that each base station is
equipped with M antennas and each user
is equipped with N antennas.
• Given every channel matrix Hkj from base
station j to user k, one can compute every
precoding matrix Vj to be used.
69/79

Non-linear Receiver with Interference
Cancellation
• Nonlinear interference cancellation (IC) may involve estimating the
interference signal at the modulation symbol level (SLIC) or at the
codeword level (CLIC) as shown in the figure.
• Error propagation issues associated with SLIC may be overcome by
adopting a soft cancellation approach, incorporating the confidence
level in estimated interference symbols. CLIC is mostly immune to
error propagation effects, but requires that the spectral efficiency
targeted by the interfering transmitter be consistent with the
interference signal quality SINR at the victim receiver.
• To successfully demodulate or decode the interference’s signal, both
approaches require knowledge of various transmission parameters of
the interfering signal.
• For SLIC, interference parameters that can enable interferer channel
estimation and interferer detection at symbol level are needed. For
CLIC, interference parameters used for interference de-scrambling and
turbo decoding are also needed, in addition to the required
interference parameters for symbol level IC.
• Interference parameters required for SLIC for a cell-center user include
the reference signal configuration, the number of data layers and the
modulation scheme. Interference parameters required for CLIC for a
cell-center user include the SLIC parameters plus the code rate, the
hybrid automatic repeat request redundancy version and the user
radio network temporary identity.
70/79

Mobility Management Challenges
• UDN will raise new challenges for mobility management
• Mobility performance will deteriorate. For instance, the rate of handover failure will
increase due to severe co-channel interference between small cells located in one
cluster. The ping-pong handover rate will increase as well.
• Mobile users may trigger frequent handovers when they move across the coverage
areas of small cells, and the signaling load, including radio resource control (RRC)
messages, X2 interface messages and signaling between core-network nodes, may
be too heavy to bear.
• To access the cell with better channel conditions, users have to take measurement of
the signaling quality of a larger number of surrounding small cells, which will
significantly increases their battery consumption.
• Potential solutions include dual connectivity, virtual cells, virtual layers,
mobility anchors and handover command diversity.
71/79

Dual Connectivity
• The architecture of dual connectivity shown
in the figure can be adopted in
heterogeneous networks to improve per-user
throughput and mobility robustness by
allowing users to be connected
simultaneously to a master cell group and a
secondary cell group via the MeNB (master
base station) and SeNB (secondary base
station), respectively.
• The control plane (C-plane) protocols and
architectures for dual connectivity assume
that there will be only one S1-MME
connection per user. The MeNB is primarily
responsible for handling the user’s RRC state.
• In dual connectivity, the SeNB and MeNB
could carry different bearers or a single
bearer split into two streams.
72/79

Virtual Cell
• A user-centric virtual cell is configured on the fly by a mobile
user at the cell center and a set of cooperative small cells
located in a circular area around them.
• The small cell in each virtual cell may vary from time to time
according to the movement of the user, or a change in the
wireless environment.
• The virtual cells, which are a cluster of cooperating hyper-
dense small cells, may need a centralized node to manage
them.
• In contrast to traditional cells, user-centric virtual cells enjoy
advantages including:
• complete elimination of cell-edge users.
• significant capacity gain from cooperative transmission or
reception.
• dynamic local cooperation, which reduces signaling overhead
and computational complexity compared with static network-
wide cooperation.
• The physical cell included in the virtual cell changes when the
user moves in the network, but the virtual cell ID remains the
same. Therefore, no handover happens while the user is
moving, and the user experience will be much enhanced. The
diagram of virtual cell changes for a moving user is shown in
the figure.
73/79

Virtual Layer
• Virtual layer technology is a solution that relies on multi-layer
networking: a virtual layer and a real layer as shown on the figure.
• The virtual layer is responsible for broadcasting, paging and
mobility management, while data transmission is carried on the
real layer.
• UDNs are divided into multiple clusters, each cluster
corresponding to one virtual layer. UEs in idle mode camp on the
virtual layer and have no need to recognize the real layer; no cell
re-selection is needed either if UEs move within the same virtual
layer. For UEs in connected mode, both the virtual layer and the
real layer can be recognized; no handover is needed for the
mobile users within the same virtual layer and so the user
experience will be good.
• Virtual layer technology can be realized by single-carrier or
multiple-carrier solutions. For the single-carrier solution, the
virtual layer and the real layer are constructed by different signals
or channels respectively. For the multiple-carrier solution, the
network can configure one carrier for each individual layer: one
for the virtual layer and one for the real layer.
74/79

Mobility Anchor
• The mobility anchor solution can reduce/hide
signaling load towards the core network by
hiding subsequent mobility involving SeNBs.
• Such a mobility anchor would be independent
of the dual connectivity solution and could
also be applied in case of limited user
capability (single RX/TX) or high system load.
• In this solution a logical entity called a
mobility anchor is introduced. The macro cell
can be used as the anchor base station for the
location of mobility anchor or the mobility
anchor can be a new entity. The potential
architectures for a mobility anchor are shown
in the figure.
• A mobility anchor is introduced as a
centralized controller or a proxy for all small
cells in one area; both S1-MME and S1-U
terminate in the mobility anchor
75/79

Handover Command Diversity
• The major reason for handover failures is
the failure of handover command
transmission for users that are out-of-
sync. Higher handover thresholds will
further increase these failures.
• One possible solution is to enable the
user to also receive the handover
command from the target cell, so that a
higher SINR can be achieved when
compared to a late handover command
transmission from the source cell.
• Moreover, if the user is able to also
receive the handover signal from the
target cell, going out-of-sync in the source
cell can be prevented.
76/79

Architecture and Backhaul
• Cloud Radio Access Network (C-RAN) is key for future mobile networks
in order to meet the vast capacity demand of mobile traffic, and
reduce the capital and operating expenditures faced by operators.
• C-RAN is a centralized, cloud-computing-based cellular network
architecture that has the ability to support current and future wireless
communication standards. It does so by separating the baseband units
(BBU) from the remote radio heads (RRH), and migrating the BBUs to
the cloud, forming a BBU pool for centralized processing. This gives
better flexibility and scalability in terms of deployment of further RRHs
compared to traditional radio access networks.
• The design of UDN involves joint consideration of various issues,
among them interference management/coordination, mobility
management, control and user plan decoupling, backhauling, as well
as multi-radio access technology (RAT) integration/interworking. In this
case, C-RAN will come to play an important role in internal high-speed,
low-latency coordination schemes and the central processing required
to implement them.
• The backhaul for UDN needs to have a significant role in aggregating
and sending data traffic between the radio access link and the
backbone network segment.
• As of today, 20% of all current backhaul deployments are copper-
based xDSL and 30% are fiber-based, microwave represents nearly
50%.
77/79

Wireless Backhaul
• Wireless backhaul will be valuable for mobile
networks, especially UDNs, due to its moderate
installation cost and relatively short deployment
time.
• Although fiber-based backhaul is better on
capacity and latency, wireless backhaul has
significant advantages over wired in terms of site
acquisition. It is also beneficial to plug-and-play
base stations.
• Therefore, a hybrid backhaul architecture, which
includes wired and wireless options, should be
reasonable for UDNs.
• The wireless mesh network (WMN), which is a
hybrid backhaul architecture, aims to construct a
high-speed, highly efficient, self-optimizing and
self-maintenance wireless transmission network
between base stations so as to fulfill the demands
for high data-transmission rates and traffic-volume
density.
78/79

Hierarchical Backhaul
• For better deployment, a WMN can be constructed with a hierarchical
backhaul architecture, where different base stations are marked by
layers.
• The first backhaul layer includes macro cells and other small cells that
have wired backhaul.
• The small cells belonging to the second backhaul layer are connected
to the base stations in the first backhaul layer with one-hop wireless
transmission.
• Likewise, the small cells belonging to the third and subsequent layers
are connected to the upper layer with one-hop wireless transmission.
• This architecture combines wired and wireless backhaul together with
an adaptive structure providing an easy and plug-and-play networking.
• A wireless point-to-point (PTP) transmission scheme can adapt load
balancing among all the small cells. Three topologies are considered:
• A tree topology with a single gateway node that connects the backhaul
network to the backbone fiber network.
• Multiple parallel gateway nodes.
• A mesh topology arrangement with redundant links.
• Different frequencies can be assigned to different parts of the
backhaul network in order to satisfy changing traffic conditions and
suppress interference.
79/79

5G Fundamentals

More Related Content

What's hot (20)

Similar to 5G Fundamentals (20)

More from Damien Magoni (8)

Recently uploaded (20)

5G Fundamentals