Cellular systems and infrastructure base wireless network

Chapter5
Cellular Systems and Infrastructure-
Base Wireless Network
北京科技大学通信工程系中山张
系方式：联 18610562032
箱：邮 zhangzs@ustb.edu.cn

 Fading and interference are the two key challenges in wireless mobile communications.
While fading has impacts on the coverage and reliability, interference affects the
reusability of spectral resource in space.
 Cellular concept was a major breakthrough in solving the problem of spectral congestion
and user capacity. It offers very high capacity in a limited spectrum allocation without any
major technological changes;
 In a cellular system, each base-station (BS) is assigned a portion of the total number of
channels available to the entire system;
 Conventionally, nearby BS’s are assigned different groups of channels, in order to mitigate
the interference from neighboring cells;
 The available channels are distributed throughout the geographic region and may be
reused as many times as necessary.
Introduction

Cellular
WIMAX
Satellite
Cellular Mobile Telephony

A
X
A
A
A
A
A A
D+R
D+R
D-R
D-R
DD
Co-channel
interferenc
e

 Structural evolution of cellular communications;
 Frequency reuse;
 Duplex techniques;
 Multiple-access/broadcasting techniques;
 Handover (handoff);
 Multi-cell cooperation/processing;
 Resource allocation;
 Cognitive radios;
 MIMO and massive MIMO;
 Distributed antenna wireless communications;
 Cellular social networks.
Summary

Figure 1: Illustration of the concept of a cell covering a given range.
Uplink
Downlink

Macrocells are large size cells, each of which can cover a radius of up to 10 miles
in diameter, depending on the terrain;
Microcells provide a mid-sized coverage, popularly employed in urban and
suburban areas. A microcell typically offers a coverage area of less than two
kilometers in diameter;
Picocells are even smaller than microcells. A picocell typically covers an area of
less than two hundred meters in diameter and are typically used for indoor
applications;
Femtocells are currently the smallest cells. Each femtocell typically covers an area
of less than 20 meters in diameter for supporting two to four simultaneous calls.
Cellular Structures

Heterogeneous networks represent the integration of Macro-,micro-, pico- and
femtocells;
It is an efficient way to increase the system capacity and improve the network
coverage;
It provides efficient ways for making use of the radio resources;
It provides integrated approaches for high-flexibility resource management;
It is capable of providing various types of services with different QoS
requirements;
According to the QoS requirements, a service may be supported by one or
simultaneously by several wireless network interfaces.
Cellular Structures: Heterogeneous Networks

SRSO - Separate resources, separate operations (conventional);
CRSO - Common resources, separate operations (based on the
techniques, such as, cognitive radios operated in the interweave or
underlay paradigm);
CRJO - Common resources, joint operations (based on the
techniques, such as, cognitive radios operated in the overlay
paradigm).
Heterogeneous Networks: Typical Operation Modes

Frequency reuse is based on the property that a wireless signal transmitted decays
with its travel distance. Hence, two wireless signals of a given frequency generate little
interference, when their transmitters are separated with a sufﬁcient distance;
In cellular wireless networks, frequency reuse allows signiﬁcant increase of capacity.
For a cellular network having in total N channels, if each cell is assigned a group of S
channels, then, the N channels can be allocated to N = N /S cells, which forms a cluster
of size N . The corresponding frequency reuse factor is 1/N .
If a cellular network has M clusters, the total number of channels of the network is
then given by C = M SN = M N .
By contrast, when a cellular network has in total M cells, the total channels of the
network is then given C = MS = MN /N , which decreases as the cluster size N increases.
Frequency Reuse

Frequency Reuse: Patterns
(a) Frequency reuse pattern for N = 4 (b) Frequency reuse pattern for N = 7
1
1
1
2
3
4
3
2 4
2
3
42
1
2
3
4
5
6
7
1
2
3
4
5
6
7
3
1
4
5
6
2
7
4
1
3
4

Frequency Reuse Pattern: N = 13
A
A
A
A
A
B
B
B
B
B
Figure 2: Frequency reuse pattern for N = 32 + 12 + 3 × 1 = 13.

 Explicitly, for obtaining the maximum capacity, we should choose cluster
size N = 1, which yields C = MN ;
In information theory, at high SNR, the capacity of a cellular system with a
frequency reuse factor 1/N is just 1/N of the capacity of a cellular system with
full frequency reuse;
However, when without using the advanced interference reduction
techniques or intelligently processing the interference, N = 1 means severe
intercell (co-channel) interference, which ultimately degrades the capacity of
the cellular networks.
 In this context, then, how do we make N close to one, but without
generating much negative impacts?
Frequency Reuse - Summary

Duplex considers the techniques (or strategies) of communications against two
directions, generally, incoming and outgoing, or uplink and downlink in cellular
wireless systems.
 FDD: frequency-division duplex;
 TDD: time-division duplex;
 CDD: code-division duplex.
 Can we use MDD (multicarrier-division duplex) and what are its
advantages and disadvantages?
Duplex

U: Uplink (incoming)
D: Downlink (outgoing)
Figure 3: Illustration of the frequency-division duplex (FDD).
Duplex: FDD

For wireless communications systems based on FDD, the uplink (incoming) and
downlink (outgoing) are separated (orthogonal) in the frequency-domain;
In FDD-assisted wireless communications, the available frequency bandwidth is
divided into two subbands, one is for the uplink transmission and the other is for
the downlink transmission, which are supported by two carrier frequencies;
The uplink and downlink subbands are separated by a so-called guard-band.
FDD: Principles

Time-Division Duplex
U: Uplink (incoming)
D: Downlink (outgoing)
Figure 4: Illustration of the time-division duplex (TDD).

TDD: Principles
For the wireless communications systems based on TDD, the uplink
(incoming) and downlink (outgoing) communications are separated
(orthogonal) in the time-domain, while communicating within the
same frequency band.
In the TDD-based wireless systems the time-axis is divided into a
number of time-slots. A time-slot can be assigned either for the
uplink (U) transmission or for the downlink (D) transmission.
Due to the fact that wireless channels experience delay-spread,
which results in ISI, a certain amount of guard-time is usually inserted
between two adjacent time-slots.

CDD: Principles
CDD is for DS-CDMA systems;
Assume there is a set of codes {ci }, which are referred to as the smart codes
and have the properties:
a.The auto-correlation coefﬁcients within a delay-window is zero or very small;
b.The cross-correlation coefﬁcients within a delay-window is zero or very small;
Then, some smart codes can be allocated to support the uplink
communications, while the rests are allocated to support the downlink
communications;
In the CDD systems, both the uplink and downlink can be operated within the
same frequency band with the aid of the TDD.

Duplex: Can be MDD?
Uplink Downlink
f0 f1 f2 f3 f4 f5 f6 f7 f8
Frequency
Figure 5: Illustration of the multicarrier-division duplex (MDD), where 1/3 of the
subbands are allocated for uplink transmission and 2/3 of the subbands are allo-
cated for downlink transmission.

MDD: Principles
When multicarrier communications, such as SC-FDMA and OFDM,
are considered, MDD may be employed for the uplink (incoming) and
downlink (outgoing) transmissions;
MDD essentially belongs to the family of FDD;
In MDD-mode both the uplink and downlink channels are operated
within the same frequency band. A fraction of the subbands
(subcarriers) can be allocated for supporting the uplink transmission,
while the others for the downlink transmission;
In MDD-mode, according to the practical requirements, the number
of subbands allocated to the uplink or downlink of a user can be ﬁxed
or dynamic. The number of subbands allocated to a user can also be
different from that allocated to another user.

Can We Use Hybrid Duplex?
 FDD+TDD - let the frequency bands for the uplink/downlink hop,
alternatively;
 TDD+MDD - the uplink transmits on one time-slot and the downlink
transmits on the other one, alternatively;
 Full-Duplex (FDX) - How far away is it from practical applications?
What are the main challenges? If cannot double the capacity, how much
can be attained?

Multiple-access/Multi-cast Techniques
In wireless communications, multiple users are supported by the so
called multiple-access/multi-cast techniques, which typically include:
 Frequency-Division Multiple-Access (FDMA): Split the channels in the
frequency domain;
 Time-Division Multiple-Access (TDMA): Split the channels in the time
domain;
 Code-Division Multiple-Access (CDMA): Using signature wave-forms
for users to transmit information in the same frequency band at the
same time;
 Space-Division Multiple-Access (SDMA): Split the channels in the
space domain.

Figure 6: Illustration of channel conﬁguration in FDMA systems. Different
users transmit signals on different frequencies at the same time.

Figure 7: Illustration of channel conﬁguration in TDMA systems. Different
users transmit signals at different time-slots using the whole frequency-band
available.

Figure 8: Illustration of channel conﬁguration in CDMA systems. Different users
are distinguished by their unique codes. All user signals are transmitted on the
same frequency-band at the same time.

Figure 9: Illustration of channel conﬁguration in SDMA/CDMA systems. Different
users or user sets can also be distinguished by their locations.

FDMA: Typical Characteristics
FDMA can support transmission of both analog and digital signals;
The frequency band supporting a FDMA system is divided into a number
of subbands, which are called as user channels;
These user channels are designed to be orthogonal in the
frequency-domain;
Each communicating user occupies one to several channels;
Subband signals usually experience ﬂat fading;
Typical examples of FDMA include classic FDMA, OFDMA, SC-FDMA, etc.

TDMA: Typical Characteristics
Single-carrier;
Time-axis is divided into the time-slots, which constitute the user
channels;
These user channels are orthogonal in the time-domain;
Each communicating user occupies one to several channels;
User signals are usually wideband signals experiencing
frequency-selective fading.

CDMA: Typical Characteristics
Each user is assigned one to several codes for signaturing its
transmitted signals;
Signature codes are expected to have good auto/cross correlation
properties;
User signals are wideband signals;
User signals usually overlap simultaneously in both frequency and time;
Can be operated either synchronously or asynchronously;
Wideband user signals, typically, experiencing frequency-selective
fading;

SDMA: Typical Characteristics
Multiple users are distinguished by their spatial signatures (channel
impulse responses);
User signals overlap simultaneously in both time and frequency;
SDMA shares most of the characteristics of CDMA;
SDMA is usually implemented associated with other multiple-access
techniques, such as FDMA, TDMA, CDMA, etc.

Handover
Handover is the procedure changing the assignment of a mobile unit from one BS
to another as the mobile moves from one cell to another:
Hard handover: A hard handover is the one in which the channel in the source
cell is released and only then the channel in the target cell is engaged. Thus the
connection to the source is broken before the connection to the target is made
(http://en.wikipedia.org/wiki/Handoff);
Soft handover: A soft handover is the one in which the channel in the source cell
is retained and used for a while in parallel with the channel in the target cell. In this
case the connection to the target is established before the connection to the
source is broken (http://en.wikipedia.org/wiki/Handoff).

Received signal
at BS A
at BS B
T h1
T h2
T h3
H
LA LBL1L2L3 L4
Figure 10: Handover decision making schemes.

Advanced Techniques for
Cellular Communication Systems
 Resource allocation;
 Multi-cell cooperation/processing (MCCP);
 Cognitive radios;
 MIMO, massive MIMO;
 Distributed antenna wireless systems;
 Cellular social networks;
 etc.

Resource Allocation
 Resources in wireless communications include
 Time;
 Space;
 Frequency spectrum;
 Power.
Resource allocation says allocating a certain amount of frequency spectrum
and a certain amount of power to transmit signals from one chosen space to
another chosen space within a given duration of time.

Resource Allocation: Degrees-of-Freedom
 In wireless communications, the resources of time, frequency and space
can in general be uniﬁed into a type of resource referred to as degrees-of-
freedom (DoFs):
 Time-domain: DoFs represent the non-overlapping time-slots;
 Frequency-domain: DoFs represent the non-overlapping channels;
 Space-domain: DoFs represent the orthogonal spatial beams.
 Then, resource allocation can be viewed as allocating the DoFs supported
by the correspondingly allocated power.

Resource Allocation
 Typical objectives of resource allocation include
 maximizing capacity (sum rate, throughput, etc.)
 maximizing reliability (minimizing error rate, maximizing SINR, etc.)
 or their joint (maximizing throughput at a given reliability, etc.)
 Resource allocation may be implemented via
 Centralized algorithms;
 Distributed algorithms.

Figure 11: An example to show the potential of using resource allocation.

 On the basis of information exchange among BS’s, MCCP can be classiﬁed into the
models:
√ CIRD-MCCP: exchange of both CIR information and data;
√ CIR-MCCP: exchange of CIR information only;
√ D-MCCP: exchange data only.
 In view of global/local information exchange among BS’s, MCCP can be classiﬁed into the
models:
ⅹ Centralized MCCP: exchange of global information;
ⅹ Distributed MCCP: exchange of local information.
 Hybrid model - formed by the combination of the above models.
MCCP: Possible Models

CIRD-MCCP - What can we do?
 A multi-cell system is equivalent to a single-cell SDMA system;
Hence, all the transmission/detection techniques for single-cell SDMA system can
be extended for the MCCP;
 A cellular system of M ideally connected BSs, each with J antennas, is capable of
supporting in total JM users, regardless of how strong the interference among them
is;
At the BSs, optimum encoding/decoding can be operated, allowing to achieve the
sum rate of multi-user MIMO systems;
 etc.

CIR-MCCP - What can we do?
Scheduling;
Coordinated power-control/allocation;
Coordinated transmitter/receiver beamforming;
Advanced coding for interference mitigation: specifically designing transmit signals
to facilitate detection at neighboring cells;
 Interference alignment: specifically designing transmit signals so that the
interferences are always constrained at the confined subspaces at each receiver, which
allows the receiver to efficiently reject the interference.
etc.

D-MCCP - What can we do?
 Uplink: Interference cancellation;
 Uplink: BS-level decode-and-forward;
 Downlink: Distributed space-time coding to achieve transmit diversity;
 Downlink: Distributed transmitter preprocessing to maximize
reliability/throughput;
 etc.

A Double-Cell Example a
H22
K users K users
H11 H21 H12
B1 B2
α α
BS Cooperation
a X. Ju, L.-L. Yang, et.al, “Spectral-efﬁciency of multicell DS-CDMA/SDMA systems with base-station co-
operation, submitted for Publication.

MIMO Equations
(1)
(2)
1 11 1 12 2 1
y H x H x n= + +
2 21 1 22 2 2
y H x H x n= + +

Spectral-Efﬁciency: Optimum Multiuser Detection
(OMUD) with Ideal BS Cooperation
(bits/s/Hz/Cell) (3)
where E [·] is with respect to H given by
(4)11 12
21 22
H H
H
H H
 
=  ÷ ÷
 
2 2 2
1 1
l og det
2
H
N
C E I HH
σ
  
= +  ÷
  

Spectral-Efficiency: Optimum Multiuser
Detection with Data Exchange
BS 1 detects as conventional, yielding the spectral-efficiency
(5)
where Σ12 denotes the covariance matrix of the interference from Cell 2 plus
the Gaussian noise.
BS 2 carries out parallel interference cancellation (PIC) before OMUD,
generating the spectral-efficiency
. (6)
In average, C = (C1 + C2 )/2 per cell.
1 2 22 222
1
log det H
C E I H H
σ
  
= +  
  
( )1
1 2 11 11 12
l og det H
N
C E I H H
− = +
  ∑

Spectral-Efﬁciency:
MMSE-MUD without BS Cooperation
where γ1 represents the SINR of a user detected by MMSE-MUD,
(8)
and RI is the covariance matrix of interference plus noise.
1
1 11,1 11,1
H
I
h R hγ −
=
( )2 1
l og 1C K E γ = × + 

MMSE-SIC without BS Cooperation
(9)
 Explicitly, the MMSE-SIC without BS cooperation is capable of achieving
the capacity of the optimum detection without BS cooperation.
( )1
2 11 11
l og det H
C E I H H
− = +
  ∑

MMSE-SIC with Data Exchange
The spectral-efﬁciency of the kth user in Cell 1 is
(10)
The per cell spectral-efficiency is
1( )
11 11 11, 11,1 0
( )
k kk H H
I i ii j j
R H H h h ψ
−
= =
= + − −∑ ∑ ∑ and by definition
0 12, 12,0, ;H
j j jh hψ ψ= =
( )
1
( k)
2 11, 11,
l og 1 , 1, 2, ,H
k k I k
C E h R h k K
−  = + = ÷   
L
1
k
k
k
C C
=
= ∑

Spectral-Efﬁciency Comparison
 Ideal Cooperation: OMUD with ideal BS cooperation;
 Single-Cell Bound: One isolate cell;
 OMUD-PIC-DE: OMUD-PIC with data exchange;
 MMSE-MUD: MMSE-MUD without BS cooperation;
 MMSE-SIC: MMSE-SIC without BS cooperation;
 MMSE-SIC-DE: MMSE-SIC with data exchange.

SpectralEfﬁciency[bits/s/Hz/Cell]
SDMA, N=8, SNR=10dB
80
70
60
50
40
30
20
10
0 0 2 4 6 8 10 12 14 16 18 20 22 24
K (Number of Users per Cell)
Figure 12: SDMA: Spectral-efﬁciency versus number of users per cell, α = 0.5.

SDMA, N=8, K=8
70
60
50
40
30
20
10
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20
SNR (dB)
Figure 13: SDMA: Spectral-efﬁciency versus SNR, α = 0.5.

SDMA, N=8, K=24
100
90
80
70
60
50
40
30
20
10
0
-4 -2 0 2 4 6 8 10 12 14 16 18 20
SNR (dB)
Figure 14: SDMA: Spectral-efﬁciency versus SNR, α = 0.5.

SDMA, SNR=10dB, K=N=8
80
70
60
50
40
30
20
10
0
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
(Intercell Interference Factor)
Figure 15: SDMA: Spectral-efﬁciency versus intercell interference strength.

MCCP: Main Challenges
Theoretic capacity of multi-cell systems, when the effect of propagation
pathloss, shadowing, fast fading are taken into account, as well as when different
information exchange schemes are considered;
Trade-off between achievable performance and the amount of information
shared among the BS’s;
 Design of efficient information exchange algorithms;
Design of the precoding/decoding algorithms that are practically reasonable,
robust and scalable;
Synchronization, channel estimation in large networks;
 Effect of delay, mobility, etc.

Cognitive Radios
 Conventional radios are regulated by fixed spectrum allocation policies, which
are operated in certain time frames, over certain frequency bands and within
certain geographical regions;
 These static spectrum assignment policies have resulted in low-efﬁciency in
usage of the precious spectrum resources;
Cognitive radios (CR) provide possible solutions to the spectrum congestion
problem by introducing opportunistic access of the licensed frequency bands that
are under-utilized;
Furthermore, CRs provide novel approaches for making efficient use of the
resources in wireless communications.

Cognitive Radios: Main Functions
Main functions of cognitive radios can be summarized as :
Spectrum sensing - determining available spectrum holes for CR users and
detecting the presence of PR users;
Spectrum management - making the efﬁciency of the available spectrum as high
as possible;
Spectrum sharing - coordinating access to the spectrum;
Spectrum mobility - maintaining seamless transition from one spectrum to
another.

Spectrum Holes: Deﬁnition a
Conventional definition - a band of frequencies that are not being used by the
primary user of that band at a particular time in a particular geographic area.
Extended definition - one to several dimensions of a hyperspace (electro-space,
transmission hyperspace, radio spectrum space or simply spectrum space, etc.) of
radio signals that are not being occupied.
Note that, the dimensions of a hyperspace may include space, temporal,
frequency, code, angle of arrival, etc.

Spectrum Holes: Classification
According to the extended definition, spectrum holes may be classified as:
Frequency holes;
Temporal holes;
 Time-frequency holes;
Space-frequency holes;
Space-time-frequency holes;
Code holes;
Angle-frequency holes;
Direction-frequency holes;

Interweave Paradigm: CR users opportunistically exploit available spectrum holes
to carry out their communications, without degrading the communication quality
of PR users.
Underlay Paradigm: CR users carry out communications along with PR users,
under the constraint that the interference caused by the CR users to the PR users
does not degrade the PR users’ communication quality;
Overlay Paradigm: Both CR and PR users carry out communications using the
same frequency spectrum in the same space. For the overlay paradigm, knowledge
to each other and cooperation between the CR and PR users are critical;
Operating Paradigms of Cognitive
Radios

fU
fU
fD
fD
A near border CR imposes severe interference
on a near boarder PR’s receiving
multiple cells
fU
fU
fU
A CR close to BS imposes severe interference
on receiving signals from a near boarder PR
CRs in FDD Cellular PR Networks
CR user
PR user
Imposes interference on

FDD Cellular PR Networks:
Characteristics
 CRs may know the locations of the PR base-station’s (BS’s); They may also know the
locations of the PR mobile terminals;
 CRs may know the PR signal’s parameters, such as frequency band, modulation, pilots,
number of active users, data rates, etc.;
 CRs may use coherent techniques to estimate the PR signals, whenever necessary;
 CRs may exploit the pilot information transmitted by the PR networks;
 CRs may cooperate with the PR BS’s or/and with the PR mobile terminals;
CRs interfere either the uplink or downlink of the PR networks;
CRs using downlink band may impose high interference on nearby mobile terminals, when
the CRs are deployed near borders of cells;

Spectrum Holes in FDD PR Systems
FDM
FDMA
TDM
TDMA
CDM
CDMA
FDD cellular PRs
Space-frequency holes
(if employs frequency reuse)
Space-frequency holes near
border for uplink band
Space-frequency holes near
BS for downlink band
Time-frequency holes when PR under-load
Space-frequency holes when a PR user and
a CR user are at diﬀerent locations, directions, etc.
Temporal holes when PR under-load
Space-time holes when a PR user and
Code holes when PR under-load
Space-code holes when a PR user and

f
f
f
f
f
ff
f
f
f
f
f
A CR user my interfere both uplink and
downlink PR users
CRs in TDD Cellular PR Networks
CR user
PR user
f
Imposes interference on
multiple users in multiple cells

TDD Cellular PR Networks: Characteristics
 CRs may know the locations of the PR BS’s; CRs may also know the locations of the PR
mobile terminals;
 CRs may know the PR transmitted signal’s parameters, such as frequency band,
modulation, pilots, number of active users, data rates, etc.;
 CRs may use coherent techniques to estimate the PR signals, whenever necessary;
 CRs may exploit the pilot information transmitted by the PR networks;
 CRs can estimate their interference on the PR users using channel reciprocity;
 CRs may cooperate with the PR BS’s or/and with the PR mobile terminals;
CRs interfere simultaneously both uplink and downlink of the PR networks;
CRs deployed near borders of the cells may impose high interference on the nearby
mobile terminals;

FDM
FDMA
TDM
TDMA
CDM
CDMA
Spectrum Holes in TDD PR Systems
Time-frequency holes when PR under-
load;
Space-frequency holes when a PR user and
TDD cellular PRs
Space-frequency holes
(if employs frequency reuse)
Temporal holes near border
during uplink transmission
Temporal holes near BS
during downlink transmission
a CR user are at different locations, directions, etc. ;
Space-time-frequency holes near border
when an uplink PR user is near BS;
Space-time-frequency holes near BS
when a downlink PR user is near border;
Temporal holes when PR under-load;
Space-temporal holes when a PR user and
a CR user are at different locations, directions, etc.;
Space-temporal holes near border
when an uplink PR user is near BS;
Space-temporal holes near BS
when a downlink PR user is near border;
Code holes when PR under-load;
Space-code holes when a PR user and
a CR user are at different locations, directions, etc.;

General MIMO and Massive MIMO
MIMO:
System model;
 Capacity of MIMO channels;
 Main challenges.
Massive MIMO:
Deﬁnition and principles;
 Main advantages;
 Main challenges.

MIMO
Main References:
1. The materials are mainly from: Yang, Lie-Liang, (2009) Multicarrier
Communications, John Wiley & Sons, Inc, Chichester, UK.
2. Cover, T.M. and Thomas, J.A., (1991) Elements of Information Theory, (New
York, USA: John Wiley & Sons, Inc).
3. Telatar, I.E., (1999) “Capacity of multiantenna Gaussian channels”, European
Trans. on Telecomm., Vol. 10, No. 6, pp. 585-595, Nov./Dec.

1
2
M
1
2
N
RX
Processor
TX
Processor
Data
Output
Data
Input
MIMO System Model
MIMO Channel: H
{h11 , h12 , . . . , h1M }
{hN 1 , hN 2 , . . . , hN M }
Figure 16: Typical representation of multiantenna MIMO systems.

MIMO: Received Signal Representation
Let us consider a MIMO system employing M transmit antennas and N receive
antennas as shown in Fig. 16. The output-input relationship of the MIMO system
can be described by the equation of
(12)
(13)
where the input and output vectors are
(14)
(15)
nHxy +=
∑=
+=
M
m
mm nxh
1
T
Mxxxx ],,,[ 21 =
T
Nyyyy ],,,[ 21 =

MIMO: Received Signal Representation
where hm represents the signature of symbol xm ;
The N -length noise vector is
(16)
(17)
The (N × M ) MIMO channel matrix is












=
=
NMNN
M
M
M
hhh
hhh
hhh
hhhH
...
...
...
],,,[
21
22221
11211
21


T
Nnnnn ],,[ 21 =

MIMO Capacity: Assumptions
The M number of symbols in x are drawn from a discrete source with zero mean
and a common variance of 1/M , i.e., E [xm ] = 0 and E [x2m ] = 1/M ;
The channels are memoryless. Each element of H obeys the complex Gaussian
distribution with mean zero and a variance 0.5 per dimension. In other words, the
channel from any transmit antenna to any receive antenna is assumed to
experience (ﬂat) Rayleigh fading;
The noise vector n is assumed to be the complex Gaussian noise vector, each
element of n is modeled as an iid complex Gaussian random variable with zero
mean and a variance of σ2 /2 = 1/2SNR per dimension, where SNR represents the
average signal-to-noise ratio (SNR) per receive antenna.

MIMO Capacity: General
Given the MIMO equation of (12) and the channel matrix H , the capacity of the
MIMO channel can be obtained by solving the optimization problem:
(18)
where
: covariance matrix of the transmitted vector x;
I (x; y | H ): mutual information between x and y , when H is given.
arg)( =HC )}|;({max
1)(:
HyxI
xQTracex ≤
][ H
x xxEQ =

Consequently, the mutual information can be expressed as:
(20)
 Finally, the capacity of the MIMO system can be obtained by solving the
optimization problem:
(21)
)]det()[(log)]det()[(log)|;( 2
2
2
2 N
NH
xN
N
IeHHQIeHyxI σπσπ −+=
]
)det(
)det(
[log 2
2
2
N
H
xN
I
HHQI
σ
σ +
=
)]
1
[det(log 22
H
xN HHQI
σ
+=
2 2: ( ) 1
1
(H) max { ( ; | } l og [ det ( )]
x
H
N xx Trace Q
C arg I x y H I HQ H
σ≤
= = +

Capacity of MIMO Channels
CSI/CSI mode: both transmitter and receiver employ channel state information
(CSI);
CDI/CSI mode: transmitter employs only channel distribution information (CDI)
and receiver employs channel state information (CSI).

MIMO Capacity: CSI/CSI Mode
When the MIMO systems are operated under the CSI/CSI mode:
both the transmitter and receiver can perfectly track the MIMO channel
matrix H ;
the transmitter can use the information about H to carry out transmitter
preprocessing, in order to achieve the capacity;
the necessary condition for achieving the capacity is that X should be chosen
to make a diagonal matrix, where Us is obtained from
sx
H
s UQU H
sss
H
UUHH ∑=

MIMO Capacity: CSI/CSI Mode
Let the rank of H be G. Then, G = min{M, N } with a probability of one, when each
element in H is an iid complex Gaussian random variable.
Let
(22)
Then, the capacity of the MIMO channels is given by
(23)
where µ is a maximal positive constant satisfying
(24)
associated with for g = 1, 2, . . . , G
},,,{
},,,{
21
21
GsS
HH
s
Gsx
H
s
diagHUHU
diagUQU
λλλ
ββββ


=∑=
==
+
=
∑==
G
g
g
HyxIHC
1
22max ][log)|;()(
σ
µλ
1)()()(
1
≤===∑=
xsx
H
s
G
g
g QTraceUQUTraceTrace ββ
+
−= )/( 2
gg λσµβ

0 5 10 15 20 25 30
0
16
14
12
10
8
6
4
2
Independent Rayleigh fading channel, CSI/CSI mode
18
(M=1, N=1)
(M=2, N=1)
(M=1, N=2)
(M=2, N=2)
(M=4, N=1)
(M=1, N=4)
Figure 17: Capacity versus SNR for the MIMO (M N ≤ 4) systems operated under the
CSI/CSI mode, when communicating over Rayleigh fading channels.
MIMOCapacity,(bits/transmission)
SNR, (dB)

0 5 10 15 20 25 30
0
25
20
Independent Rayleigh fading channel, CSI/CSI mode
30
(M=12, N=1)
(M=1, N=12)
(M=6, N=2)
(M=2, N=6)
(M=4, N=3)
(M=3, N=4)
15
10
5
SNR, (dB)
Figure 18: Capacity versus SNR for the MIMO (M N = 12) systems operated under the
CSI/CSI mode, when communicating over Rayleigh fading channels.

048
M
0
3224
Capacity (bits/transmission)
35
30
25
20
15
10
5
0
28 16
N
20
124
168
20
32 28 24
12

CSI/CSI Mode: Observations
 The capacity surface is symmetric in terms of M and N , which suggests that:
The capacity of the MIMO system using M transmit antennas and N receive
antennas is the same as that of the MIMO system using N transmit antennas and
M receive antennas, when the MIMO system is operated under the CSI/CSI mode.

MIMO Capacity: CDI/CSI Mode
For the MIMO systems operated under the CDI/CSI mode:
The receiver employs ideal knowledge about H ;
The transmitter knows only the MIMO channel’s distribution information;
Hence, the transmitter can only design the transmitted signals using the MIMO
channel’s distribution information;
The transmitted signal vector x is hence independent of the MIMO channel matrix
H ;

Proved by Telatar that, in order to achieve the capacity under the CDI/CSI mode,
the transmitted signal vector x should be circularly symmetric complex Gaussian
with zero mean and a covariance matrix
Correspondingly, the ergodic capacity of the MIMO channels under CDI/CSI mode
is
[bits/transmission] (25)
[bits/transmission]
(26)
Mx I
M
Q
1
=
)]}
1
[det({log 22 HH
M
IEC H
MH
σ
+=
)]}
1
[det({log 22
H
NH HH
M
IE
σ
+=

Special Case 1: Capacity of SISO
For a memoryless SISO system, the capacity is given by
(27)
where
represents the SNR;
h is the normalized complex gain of the wireless channel.
)||
1
1(log 2
22 hC
σ
+=
)||1(log 2
2 hγ+= [bits/transmission]
2
/1 σγ =

Special Case 2: Capacity of SIMO
For a memoryless SIMO system the capacity is given by
where
hn : the normalized complex gain of the channel associated with the nth
receive antenna;
N : the number of receive antennas;
Maximal ratio combining (MRC) based detection : optimum and achieves the
capacity.
)]||
1
1([log
1
2
22}{ ∑=
+=
N
n
nh hEC n
σ

Special Case 3: Capacity of MISO
For a MISO system, the capacity is given by
[bits/transmission]
where
hm : the normalized complex gain with respect to the mth transmit antenna;
M : the number of transmit antennas;
Open-loop optimum transmitter coding : required for achieving the capacity.
)]||
1
1([log
1
2
22}{ ∑=
+=
M
m
mh h
M
EC m
σ
(29)

Special Case 4: N is ﬁx, M → ∞
When in (26) N is ﬁxed, by the law of large number, we have
(30)
with probability one.
In this case,
which shows that the capacity of the MIMO system increases linearly with the number of
receive antennas.
)]}
1
[det({loglim 22
H
NH HH
M
IEC
σ
+=
)
1
1(log
)]
1
[det(log
22
22
σ
σ
+×=
+=
N
II NN
N
H
M
IHH
M
=
∞→
1
lim

Special Case 5: M is ﬁx, N → ∞
When in (25) M is ﬁxed, by the law of large number, we have
(32)
In this case,
[bits/transmission]
which shows that the capacity of the MIMO system increases at least linearly with the
number of transmit antennas.
)]}
1
[det({loglim 22 HH
M
IEC H
MH
N σ
+=
∞→
M
H
M
IHH
N
=
∞→
1
lim
)
1
1(log)][det(log
)]}
1
[det({loglim
2222
22
σσ
σ
+×≥+=
+=
∞→
MI
M
N
I
HH
M
IE
MM
H
MH
N
(33)

In general, if both the number of transmit antennas M and the number of
receive antennas N simultaneously become large, the capacity of the MIMO
system then grows at least linearly with G = min(M, N ).

If the value of M = N is sufficiently high, we have
(34)
Hence, when M = N → ∞, the capacity of the MIMO system satisfies
(35)
Therefore, when the values of both M and N are sufficiently high, the capacity of
the MIMO system increases at least linearly with the SNR value.
)
1
det()
1
det( 22 MM
H
M I
M
IHH
M
I
σσ
+≥+
)]}
1
[det({loglim 22 HH
M
IEC H
MH
M σ
+=
∞→
eSNRe
M
I
M
IE
M
M
MMH
M
22222
22
loglog
1
])
1
1[(loglim
)]}
1
[det({loglim
×==+=
+≥
∞→
∞→
σσ
σ
Special Case 6: M = N → ∞

Independent Rayleigh fading channel, CDI/CSI mode
Figure 19: Capacity versus SNR for the MIMO ( ) systems operated
under the CDI/CSI mode, when communicating over Rayleigh fading channels.
4≤MN

Independent Rayleigh fading channel, CDI/CSI
SNR, (dB)
Figure 20: Capacity versus SNR for the MIMO (M N = 12) systems operated
under the CDI/CSI mode, when communicating over Rayleigh fading channels.

Figure 21: Capacity versus the number of transmit/receive antennas for the MIMO
systems operated under the CDI/CSI mode, when communicating over Rayleigh
fading channels with = SNR = 1.
2
σ

CDI/CSI Mode: Further Observations
 Once the number of transmit antennas reaches the number of receive antennas,
further increasing the number of transmit antennas only results in marginal increase of
capacity.
 The reason is that, once M ≥ N , using the approximation of H H H
/M ≈ I N we obtain
(36)
 which suggests that the capacity of the MIMO system retains nearly constant,
 once the number of transmit antennas is sufﬁciently high.






+=











+=


















+≈


















+=
2222
22
22
1
1log
1
detlog
1
detlog
11
detlog
σσ
σ
σ
NII
IIE
HH
M
IEC
NN
NNH
H
NH

Figure 22: Capacity versus the number of transmit/receive antennas for the MIMO
systems operated under the CDI/CSI mode, when communicating over Rayleigh
fading channels with = SNR = 1.
2
σ

CDI/CSI Mode: Further Observations
 When the number of receive antennas exceeds the number of transmit antennas, the
capacity of the MIMO system increases more or less following the logarithm law;
 The reason is that, if N > M , using the approximation of H H
H /N = I M, we have
(37)
 Hence, for a ﬁxed value of M , the capacity of the MIMO system increases with the
logarithm of N representing the number of receive antennas.






+=











+=


















+≈


















+=
2222
22
22
1logdetlog
detlog
1
detlog
σσ
σ
σ
M
N
MI
M
N
I
I
M
N
IE
HH
NM
N
IEC
MM
MMH
H
MH

048
M
0
3224
30
25
20
15
10
5
0
28 16
N
20
124
168
20
32 28 24
12

CDI/CSI Mode: Observations
The capacity surface is asymmetric in terms of M and N , which
suggests that:
The capacity of the MIMO system using M transmit antennas and N receive
antennas is not the same as that of the MIMO system using N transmit antennas
and M receive antennas;
 Given M > N , the system using M transmit antennas and N receive antennas
may achieve signiﬁcantly smaller capacity than the system using N transmit
antennas and M receive antennas.

 If M and N simultaneously become large, the capacity of the MIMO system
grows linearly with G = min(M, N );
 The linearly growing capacity is achieved, when communicating over a rich
scattering environment providing independent transmission paths from each
transmit antenna to each receive antenna;
 This characteristics of linearly growing capacity is retained, provided
that the receiver employs the channel state information, while the
transmitter employs either the channel state information (CSI) or channel
distribution information (CDI);
MIMO Capacity - Conclusions

MIMO Capacity - Conclusions
…
 When the MIMO system employs multiple transmit antennas and when
the number of receive antennas is relatively low, such as when N ≤ M , the
capacity of the MIMO system operated under the CSI/CSI mode can be
signiﬁcantly higher than the capacity of the MIMO system operated under
the CDI/CSI mode;
 When the number of receive antennas is signiﬁcantly higher than
the number of transmit antennas, ie., when N >> M , the capacity of the
MIMO systems under both the CSI/CSI and CDI/CSI modes is similar;
 Hence, when operated under the CDI/CSI mode, it is de- sirable to use
more receive antennas, when M N is a constant.

Massive MIMO: References
1. P. Judge, “LTE may make way for massive MIMO,” in
http://www.techweekeurope.co.uk/interview/lte-may-make-way-
for-massive-mimo-7376, 2010.
2. J. Hoydis, S. ten Brink, and M. Debbah, “Massive MIMO: How
many antennas do we need?,” in The 49th Annual Allerton
Conference on Communication, Control, and Computing (Allerton),
pp. 545–550, IEEE, 2011.
3. J. Jose, A. Ashikhmin, T. Marzetta, and S. Vishwanath, “Pilot
contamination and precoding in multi-cell TDD systems,” IEEE
Transactions on Wireless Communications, vol. 10, pp. 2640 –
2651, August 2011.
4. F. Rusek, D. Persson, B. K. Lau, E. G. Larsson, T. L. Marzetta, O.
Edfors, and F. Tufvesson, Scaling up MIMO: Opportunities and
Challenges with Very Large Arrays, IEEE Signal Proces. Mag., to
appear, 2012.

M=16
M=8
M=4
M=2
M=1
40
30
20
10
0
Capacity of MIMO channels with M=N
80
M=32
70
60
50
0 1 2 3 4 5 6 7 8
SNR ( γ )
Massive MIMO,
Why?

 A system has a huge number, such as one to several hundreds, of antenna
elements;
 The number of terminals simultaneously supported is not limited by the
number of antenna elements, but, instead, limited by the incapability to
acquire the necessary knowledge for supporting the system;
 The number of antenna elements (or DoFs) is typically (or a least) an order
higher than the number of terminals supported.
Massive MIMO: Concepts [4]

Assume that a TDD-based cell, whose BS has N antennas, uses the massive
MIMO principles to support K mobile terminals, each of which employs one
antenna.
Reverse Link :
Forward Link :
yr = Hxr + nr
yf = HT xf + nf
(38)
(39)
where
 N >> K , typically, N > 10K ;
 yr , nr : (N × 1) complex vectors; yf , nf : (K × 1) complex vectors;
 xr : (K × 1) complex vector; xf : (N × 1) complex vector;
 H = [h1 , h2 , · · · , hK ]: (M × K ) propagation matrix.
Massive MIMO: System Model

 Columns of H become (nearly) orthogonal ;
 No user cooperation is required to achieve the sum-rate that is achievable when the
users are in cooperation;
 Linear processing for transmission and detection, such as TMRC and MRC, is
optimum;
 Background noise can be averaged out , the average SNR attainable
increases as the number of antennas increases;
 The achievable performance is not much related to communication channels, owing to
the huge diversity .
 The randomness in conventional MIMO becomes deterministic;
 Performance of massive MIMO is robust, the failure of some antenna elements would
not result in much performance degradation.
( )1hh k
H
k →
( )0nh r
H
k →
k
H
IHH >−
Massive MIMO: Main Characteristics

 Antenna correlation: Given the size of an antenna array, the antenna
elements become more correlated as the number of elements increases;
 Pilot contamination: The pilot signals of one cell (Cell A) are polluted by the
pilot signals from the neighboring cells. Consequently, the transmitted vectors
from the BS of cell A will be partially focus on the terminals in the neighboring
cells;
 Consequently, the system is entirely limited from the reuse of pilots in
neighboring cells;
Massive MIMO: Main Challenges

 One of the important lessons learned from cellular systems is that the cell size
should be reduced, so that the limited spectrum
 resource can be re-used by more small sized spaces, in order to increase the
system capacity;
 Consequently, when covering a give area, more and more antennas can be
deployed to divide the space into many sub-spaces;
Distributed Antenna Cellular Concept

120o
Figure 23: When each cell is divided into three sectors, three times of capacity
may be attained.

Figure 24: The capacity of using 7 distributed antennas in each cell can be much
higher than that of each BS using 7 co-located antennas

Figure 25: Can we use the distributed antenna systems, where each mobile user
is the center of a cell?

 High capacity;
 Low-power communications;
 No power-control is necessary;
 No handoff needs to be considered;
 High-robustness to failure of some antennas;
 It is a type of massive systems with distributed processing.
Distributed Antenna Cellular Systems: Possible
Advantages

 Today, mobile devices have been integrated into our daily lives;
 Mobile social networks (MSNs) are the mobile communication systems which involve
social relationship of mobile devices;
 In MSNs, mobile users can access, share, and distribute data in mobile environments by
exploiting social relations;
 MSNs belong to a class of delay tolerant networks (DTNs) that can take advantage of
human interaction and physical mobility;
 In MSNs, the social aspects (behaviors) of mobile devices can be exploited in the context
of information and communication technologies to improve the efﬁciency of data exchange,
sharing, delivery services, etc.
Cellular Social Networks: Introduction

Mobile Social Networks: Small-World
Figure 26: The “six degrees of separation”model.
(http://en.wikipedia.org/wiki/Small-world-experiment)

 Frequency and duration of encounter (contact);
 Friendship of two mobile devices;
 Community;
 ‘Small-world’ phenomenon.
 Popularity (connectivity) of a mobile device;
 Relationship of one mobile device with the other mobile devices;
Social Aspects in Social Networks

Mobile Social Networks: Friendship
T
t
0
T
t
0
T
t
0
T
t
0
(a)
(b)
(c)
(d)
∆ta
∆tb
∆tc
∆td
Figure 27: Cases of two people meeting with each other during [0, T ].

 Routing;
 Content distribution;
 Coverage extension and intercell interference mitigation in cellular
mobile systems.
 Communication in rural areas;
 Emergency communication;
 One laptop per child;
Mobile Social Networks: Applications

Conventional Cellular Systems: An
Example
f1
f2
f1
f3

Conventional Double-Cell Cellular
Systems: Characteristics
 Frequency band f1 is used for supporting the users near BSs;
 In order to mitigate the intercell interference, frequency bands f2 and f3 are
assigned to the edge users of the left and right cells;
 The total bandwidth is f1 + f2 + f3 and the frequency reuse factor is in [1/2, 1],
depended on the relative bandwidths of f1 , f2 and f3 ;
 There exists trade-off between the frequency reuse factor and intercell
interference: intercell interference increases, as the frequency reuse factor tends
to one.
 Furthermore, in order to guarantee the edge users’ QoS, BSs may need to
radiate possibly very high power.

Cellular Social Networks: An example

 Users are divided into the active subscribers (ASs), which may
communicate with both BSs and other mobile users, and the passive
subscribers (PSs), which only communicate with mobile users.
 Content distribution is completed via two steps:
 ASs ﬁrst receive the content from BSs;
 Contents are distributed by the ASs as well as the PSs that
have obtained the content, until all the mobile users receive
the content.
 The strategy may be modiﬁed to consider the other rules, such as, that a
mobile user can turn to a BS to obtain the content, if it cannot get the content
from other mobile users.
Cellular Social Networks: Operations

 One frequency band can be used for BSs to convey content to ASs, and
another frequency band can be used for content distribution within mobile users.
In this case, total two frequency bands are required, in comparison to three
required by the conventional scheme;
 In FDD systems, content distribution among mobile users may be operated on
the uplink frequency band;
 In either way, zero intercell interference is possible;
 Furthermore, as BSs only communicate with the users close to them, low BS
transmit power is attainable;
 Alternatively, cell size covered by BSs can be extended.
Cellular Social Networks:
Characteristics

 Technology: novel techniques for physical layer, network layer, etc., energy-
efficient algorithms for mobile devices, novel resource allocation algorithms, etc.;
 Good models that can closely model mobility patterns and social aspects;
 Techniques dealing with different services with different QoS requirements;
 Joint optimization algorithms that are capable of taking into account of
different social aspects;
 Cross-network optimization algorithms that can efficiently and simultaneously
consider MSNs and other structured/unstructured wireless networks;
 Algorithms dealing with selfishness and fairness;
 Standardization.
MSNs: Challenges

 Wireless communications systems without cells;
 Wireless networking without layers;
 No duplexing for up/down-links;
 Cognitive radioing without primary/secondary users;
 Wireless devices without limit on spectrum-access;
 Wireless communications not just using radio signals;
Perspectives of Future
WComms

Chapter15 Cellular Systems and Infrastructure-Base Wireless Network
Word subscribers:4300million ; Chinese subscrbers:640million
Worldwide Telecom Statistics

Cellular systems and infrastructure base wireless network

First Mobile
Radio
Telephone
1924

A
C
E
D
B
G
F
C
E
D
B
G
F
A
C
E
D
B
G
F
A
Cellular Mobile Telephony
Reuse factor
is1/7
Frequency modulation
Antenna diversity
Cellular concept
Bell Labs(1957&1960)
Frequency reuse
Typically every 7 cells
Handoff as caller moves
Modifies CO switch
HLR, paging, handoffs
Sectors improve reuse
Every3 cells possible

2
7
3 1
6
5
4
Sectoring
Cellular concept
Bell Labs(1957&1960)
Frequency reuse

A
B
C
D
E
F
G
B
C
D
E
F
G
Cell splitting
Cellular concept
Bell Labs(1957&1960)
Frequency reuse

1st
Generation Analog Cellular Systems
Standard Region Frequency
(MHz)
Channel
Spacing
(kHz)
No. of
Channels
Modulation Data Rate
(kbps)
AMPS USA 824-849
869-894
30 832 FM 10
TACS Europe 890-915
935-980
25 1000 FM 8
ETACS UK 872-915
917-950
25 1240 FM 8
NMT 450 Europe 453-457.5
463-467.5
25 180 FM 1.2
NMT 900 Europe 890-915
935-960
12.5 1999 FM 1.2
C-450 Germany
Portugal
450-455.74
460-465.74
10 573 FM 5.28
RTMS Italy 450-455
460-465
25 200 FM -
Radiocom
2000
France 414.8-418
424.8-428
12.5 250 FM -
NTT Japan 870-885 25 600 FM 0.3
JTACS /
NTACS
Japan 860-870
915-925
25 400 FM 8.0

2nd
Generation Cellular and Cordless Systems
System
Country
IS-54
USA
GSM
Europe
IS-95
USA
CT-2
Europe,
Asia
CT-3
DCT-90
Sweden
DECT
Europe
Access
Technology
TDMA /
FDMA
TDMA /
FDMA
CDMA /
FDMA
(DS)
FDMA TDMA /
FDMA
TDMA /
FDMA
Frequency
Band
BS(MHz) 869-894 935-960 869-894 864-868 862-866 1800-1900
MS(MHz) 824-849 890-915 824-849
Duplexing FDD FDD FDD TDD TDD TDD
RF Channel
Spacing
(kHz)
30 200 1250 100 1000 1728
Modulation Pi/4
DQPSK
GMSK BPSK /
QPSK
GFSK GFSK GFSK
Frequency
Assignment
Fixed Fixed Fixed Dynamic Dynamic Dynamic
Power
Control
MS Y Y Y N N N
BS Y Y Y N N N
Speech
Coding
VSELP RPE-LTP QCELP ADPCM ADPCM ADPCM
Speech rate
(kbps)
7.95 13
8
(variable
rate) 32 32 32
Channel Bit
Rate (kbps) 48.6 270.833 1228.8 72 640 1152
Channel
Coding
1/2 rate
convolution
1/2 rate
convolution
1/2 rate
forward,
1/3 rate
reverse,
CRC
None CRC CRC

BSS
HLR
AuC
C, D
Gw-MSC C
E,ISUP
PSTN/ISDN
ISUP
GSM
04.08
Call
MSC
VLR
A
UE
SMS-GW
Billing
Center
SCP
STP
IN
gsm
SCFSSP
Circuit domain
GSM Network

GSM & GPRS
BSS
HLR
AuC
C, D
Gw-MSC
C
E,ISUP
PSTN/ISDN
ISUP
GSM
04.08+
Call
MSC
VLR
A
UE
SMS-GW
Billing
Center
GGSN
PDN
Gi
Gb
SGSN
Data,
voice,
video
call
GSM
04.08+
Gr
Gc
Gn
CGw
Ga
Ga
SCP
STP
IN
gsm
SCFSSP IP Services
Circuit domain Packet domain

GPRS
General Packet Radio Service
 Packet based Data Network
 Well suited for non-real time internet usage including retrieval of email, faxes
and asymmetric web browsing.
 Supports multi user network sharing of individual radio channels and time
slots.
 Provides packet network on dedicated GSM radio channels
 GPRS overlays a packet-switched architecture on existing GSM network
architecture
Variable performance…
 Packet Random Access, Packet Switched
 Content handling
 Throughput depends on coding scheme, # timeslots etc
 From ~ 9 kbps min to max. of 171.8 kbps (in theory!)

GPRS (contd..)
 Modulation – GMSK
 Symbol Rate – 270 ksym/s
 Modulation bit rate – 270 kbps
 Radio data rate per time slot – 22.8kbps
 User data rate per time slot – 20kbps (CS4)
 User data rate (8 time slots) – 160kbps, 182.4kbps
 Applications are required to provide their own error correction scheme
as part of carried data payload.

GSM evolution to 3G
GSM
9.6kbps (one timeslot)
GSM Data
Also called CSD
GSM
General Packet Radio Services
Data rates up to ~ 115 kbps
Max: 8 timeslots used as any one time
Packet switched; resources not tied up all the time
Contention based. Efficient, but variable delays
GSM / GPRS core network re-used by WCDMA (3G)
GPRS
HSCSD
High Speed Circuit Switched Data
Dedicate up to 4 timeslots for data connection ~ 50 kbps
Good for real-time applications c.w. GPRS
Inefficient -> ties up resources, even when nothing sent
Not as popular as GPRS (many skipping HSCSD)
EDGE
Enhanced Data Rates for Global Evolution
Uses 8PSK modulation
3x improvement in data rate on short distances
Can fall back to GMSK for greater distances
Combine with GPRS (EGPRS) ~ 384 kbps
Can also be combined with HSCSD
WCDMA

 CS1 guarantees connectivity under all conditions (signaling and start of data)
 CS2 enhances the capacity and may be utilised during the data transfer phase
 CS3/CS4 will bring the highest speed but only under good conditions
Channel data rates determined by Coding Scheme
3dB7dB11dB15dB19dB23dB27dB C/I
0
4
8
12
16
20
MaxthroughputperGPRSchannel
(nettobitrate,kbit/sec)
CS 4
CS 3
CS 2
CS 1
Use higher coding schemes (less coding, more payload) when radio conditions are
good

 EDGE Enhanced Data Rates for Global Evolution
EDGE is add-on to GPRS
Uses 8-PSK modulation in good conditions
Increase throughput by 3x (8-PSK – 3 bits/symbol vs GMSK 1 bit/symbol)
Offer data rates of 384kbps, theoretically up to 473.6kbps
Uses 9 Modulation coding schemes (MCS1-9)
MCS(1-4) uses GMSK, while MCS(5-9) uses 8PSK modulation.
Uses Link adaptation algorithm
Modulation Bit rate – 810kbps
Radio data rate per time slot – 69.2kbps
User data rate per time slot – 59.2kbps (MCS9)
User data rate (8 time slots) – 473.6kbps
 New handsets / terminal equipment; additional hardware in the BTS, Core network and the rest
remains the same
 EDGE access develops to connect to 3G core
EDGE

WCDMA/UMTS
UTRAN
HLR+
AuC
C+, D+
Gw-MSC C
E+,ISUP
PSTN/ISDN
ISUP
GSM
04.08++
Call
3G-MSC
VLR
Iu-cs
UE
SMS-GW
Billing
Center
GGSN
PDN
Gi+
Iu-ps
3G-SGSN
Data,
voice,
video
call
GSM
04.08++
Gr+
Gc+
Gn+
CGw
Ga+
Ga+
SCP
STP IN, CAMEL
gsm
SCFSSP IP Services
Circuit domain Packet domain

UMTS
 UMTS is the European vision of 3G.
 UMTS is an upgrade from GSM via GPRS or EDGE.
 The standardization work for UMTS is carried out by Third Generation
Partnership Project (3GPP).
 Data rates of UMTS are:
144 kbps for rural
384 kbps for urban outdoor
2048 kbps for indoor and low range outdoor
 Virtual Home Environment (VHE)

UMTS Network Architecture
Mobile Station
MSC/
VLR
Base Station
Subsystem
GMSC
Network Subsystem
AUCEIR HLR
Other Networks
Note: Interfaces have been omitted for clarity purposes.
GGSN
SGSN
BTS
BSC
Node
B
RNC
RNS
UTRAN
SIM
ME
USIM
ME
+
PSTN
PLMN
Internet

GSM/UMTS Bit rate, Mobility and Services

IMT-2000 Vision Includes LAN, WAN and Satellite Services

 Higher bandwidth enables a range of new applications!!
 For the consumer
Video streaming, TV broadcast
Video calls, video clips – news, music, sports
Enhanced gaming, chat, location services…
 For business
High speed teleworking / VPN access
Sales force automation
Video conferencing
Real-time financial information
Why 3G?

3G services in Asia
CDMA (1xEV-DO)
 Korea: SKT, KTF
 Japan: AU (KDDI)
WCDMA / UMTS
 Japan: NTT DoCoMo, Vodafone KK
 Australia: 3 Hutchinson
 Hong Kong: 3 Hutchinson

3G Standards
 3G Standard is created by ITU-T and is called as IMT-2000.
 The aim of IMT-2000 is to harmonize worldwide 3G systems to provide Global
Roaming.

IS-95
2G 2.5G 3G
GSM
Is-136
&PDC
IS-95B
HSCSD
GPRS
EDGE
CDMA2000-
1XRTT
W-CDMA
EDGE
CDMA2000-
1XREV,DV,DO
CDMA2000-
3XRTT
3GPP2
TD-SCDMA 3GPP
Upgrade paths for 2G Technologies

cdmaOnecdmaOne
GSMGSM
TDMATDMA
2G
PDCPDC
CDMA2000
1x
CDMA2000
1x
First Step into 3G
GPRSGPRS 90%
10%
EDGEEDGE
WCDMAWCDMA
CDMA2000
1x EV/DV
CDMA2000
1x EV/DV
3G phase 1 Evolved 3G
3GPP Core
Network
CDMA2000
1x EV/DO
CDMA2000
1x EV/DO
HSDPAHSDPA
Expected market share
EDGE
Evolution
EDGE
Evolution
- drivers are capacity, data speeds, lower cost of delivery for revenue growth
Evolution of Mobile Systems to 3G

2G 3G
and Beyond
IP
Evolution from 2G
systems
Revolution from subscriber
service expectations
Revolution from subscriber
service expectations
Come from IP

Performance evolution of cellular technologies

Improved performance, decreasing cost of delivery
Typical
average bit
rates
(peak rates
higher)
WEB browsing
Corporate data access
Streaming audio/video
Voice & SMS Presence/location
xHTML browsing
Application downloading
E-mail
MMS picture / video
Multitasking
3G-specific services take
advantage of higher bandwidth
and/or real-time QoS
3G-specific services take
advantage of higher bandwidth
and/or real-time QoS
A number of mobile
services are bearer
independent in nature
A number of mobile
services are bearer
independent in nature
HSDPA
1-10
Mbps
WCDMA
2
Mbps
EGPRS
473
kbps
GPRS
171
kbps
GSM
9.6
kbps
Push-to-talk
Broadband
in wide area
Video sharing
Video telephony
Real-time IP
multimedia and games
Multicasting
Services roadmap
CDMA
2000-
EVDO
CDMA
2000-
EVDV
CDMA
20001x

Drawbacks of previous generation
 1G compares unfavorably to its successors. It has low capacity, unreliable handoff,
poor voice links, and no security at all since voice calls were played back in radio
towers, making these calls susceptible to unwanted eavesdropping by third parties.
 2G technologies weaker digital signals may not be sufficient to reach a Cell tower.
 2G Difficult roaming between countries using different systems.
 Back ground Noise, lossy compression during CODECS.
Need of 3G

 3G wireless technology represents the convergence of various 2G wireless
telecommunications systems into a single global system that includes both
terrestrial and satellite components.
 3G High-speed, mobile supports video and other rich media, always-on
transmission for e-mail, Web browsing, instant messaging.
 It is based on the International Telecommunication Union (ITU2000) family of
standards Services include wide-area wireless voice telephony, video calls, and
broadband wireless data, all in a mobile environment.
What is New in 3G?

 Global Roaming.
 Send and Receive E-Mail Messages.
 High Speed Web.
 Superior Voice Quality.
 Tele/Video Conferencing.
 Electronic agenda meeting reminder.
 3d Animation Games.
 Website creating Using Mobile Phones.
 Etc….
The Features of 3G

Our Real Time Implementation
in 3g Technology
…………….

• Both Remote and Local area
Students can easily get interact with
queries, and listen at the same time.
• Access in
Inside campus -
Remote areas -
WI-FI
Wi-Max

WI-FIWI-FI
Block 2
Block 3
Block 1
Inside Campus

Wi-MAXWi-MAX
Area 2
Area 3
Area 1
both
Remote areas and
Local area

Remote area Local area
Open Smart Class rooms

Features of Implementation
 Both Local area and Remote areas Students
Can interact the teacher by
 Queries
 Suggestions
 Feedback
 Language Translation.
 World wide access.

2G & 3G — CDMA
Code Division Multiple Access
Spread spectrum modulation
 Originally developed for the military
Resists jamming and many kinds of interference
Coded modulation hidden from those w/o the code
All users share same (large) block of spectrum
 One for one frequency reuse
Soft handoffs possible
Almost all accepted 3G radio standards are based on CDMA
CDMA2000, W-CDMA and TD-SCDMA

W-CDMA : makes possible a world of mobile multimedia

WCDMA Domains
User Equipment
Domain
Access
Network
Domain
Core
Network
Domain
Infrastructure
Domain
Cu
Mobile
Equipment
Domain
USIM
Domain
Home
Network
Domain
Transit
Network
Domain
Uu Iu
[Zu]
[Yu]
Serving
Network
Domain
Standardization of architecture (domains) and standardization of protocols (strata)

WCDMA Protocol Layers
Application
Protocol
Data
Stream(s)
ALCAP(s)
Transport
Network
Layer
Physical Layer
Signalling
Bearer(s)
Transport
User
Network
Plane
Control Plane User Plane
Transport
User
Network
Plane
Transport Network
Control Plane
Radio
Network
Layer
Signalling
Bearer(s)
Data
Bearer(s)

WCDMA L1, L2, and RRC Sublayer
L3
con
trol
con
trol
con
trol
con
trol
Logical
Channels
Transport
Channels
C-plane signalling U-plane information
PHY
L2/MAC
L1
RLC
DCNtGC
L2/RLC
MAC
RLC
RLC
RLC
RLC
RLC
RLC
RLC
Duplication avoidance
UuS boundary
BMC L2/BMC
RRC
control
PDCP
PDCP L2/PDCP
DCNtGC
L3/RRC

Logical Channels Control Traffic
BCCH PCCH DCCH CCCH SHCCH DTCH CTCH
Mac -b -c/sh -d
Common Dedicated
Transport Channels BCH PCH FACH RACH UL CPCH DSCH DCH
Physical Channels Mapped to Transport Channels Dedicated
PCCPH SCCPCH PRACH PCPCH PDSCH DPDCH DPCCH SCH
CPICH
AICH
PICH
CSICH
CD/CA-ICH
Transport Channels: how information transferred over the radio interface
Logical Channels: Type of information transferred over the radio interface
Channels made by soft hats
WCDMA Channels

Mapping Between Channels
SCH
CPICH
AICH
PICH
CSICH
CD/CA-ICH
CCCH
DCCH
DTCH PCCH BCCH CCCH CTCH
DCCH
DTCH
RACH CPCH DCH PCH BCH FACH DSCH DCH
Logical
Channels
Transport
Channels
Uplink Downlink
PCCPCH SCCPCHPRACH DPDCH
DPCCH
PDSCHPCPCH
Mapped
Physical
Channels
Dedicated
Physical
Channels
DPDCH
DPCCH
N to M

WCDMA Channel Usage Examples
Dedicated channels Common channels Shared channels
DCH FCH RACH CPCH DSCH USCH
Uplink/ Both Downlink Uplink Uplink Downlink Uplink, only
Downlink in TDD
Code Usage According to maxm Fixed Fixed Fixed Codes Codes
bit rate codes per codes per codes per shared shared
cell cell cell btw users btw users
Fast Power control Yes No No Yes Yes No
Soft handover Yes No No No No No
Suited for Medium or large Small Small Small or Medium Medium
data amounts data data medium or large or large
amounts amounts data data data
amounts amounts amounts
Suited for bursty No Yes Yes Yes Yes Yes
data
Flexibility comes with responsibility

Radio Resource Management
Power Control
Handover
Access Control
Load and Congestion Control
Packet Scheduling

WCDMA Power Control (near = far)
YY
NodeB
Keep received power
levels P1 and P2 equal
Power control commands
to the UEs
UE1
UE2
Uplink and downlink (1500 Hz)
Open Loop Power Control
Closed Loop Power Control
Outer Loop Power Control
Equal Opportunity Administration (EOA)

WCDMA Handovers
YY
sector 1
sector 2
RNC
The same signal is sent
from both sectors to UE
RNC
YY
YY
NodeB1
NodeB2
The same signal is sent from
both NodeB's to UE, except for the
power control commands
macro diversity
combining in uplink
Hard and Inter-frequency handovers
Intersystem cell-reselection
“Equivalent PLMN mode” (autonomous cell re-selection (packet) idle mode)
Softer
Soft

BS1
BS2
A B
Time
Time
Level at point A
Level at point B
Handoff threshold
Minimum acceptable signal
to maintain the call
Level at point B(call is terminated)
Level at which handoff is made
(call properly transferred to BS2)
(a) Improper
Handoff
situation
(b) proper
Handoff
situation
READY_TO_SW
ITCH_IN
ACK_TO_S
WITCH_IN
DL
Transmission
UL Transmission

Small microcells for
low speed traffic
Large “umbrella” cell
for high speed traffic

Handover Algorithm
Pilot Ec/IO of cell 1
Reporting_range
- Hysteresis_event 1A
T T T
Reporting_range
+ Hysteresis_event 1B
Hysteresis_event 1C
Connected to cell 1
Event 1A
- add cell2
Event 1C
= replace cell1
with cell3
Event 1B
= remove cell3
A relay race with multiple batons

Dimensioning Criteria
—Coverage, Capacity, Quality of Service
Dimensioning
—Link budget, capacity (hard and soft) and load factor
—Estimation of average interference power
—Coverage end Outage probabilities
Optimization
—Performance Requirements
—Antenna adjustments, neighbor lists, scrambling codes
Don’t force a round peg in a square hole
Network Dimensioning and Optimization

WCDMA Quality of Service (Qos)
 Dynamic Negotiations of properties / Services of radio bearer
—Thruput, transfer delay, data error rate
—Authentications
Traffic class Conversational class Streaming class Interactive class Background
Fundamental Preserve time relation Preserve time Request response Destination is not
characteristics (variation) between relation (variation) pattern expecting the data
information entities of between information Preserve data within a certain time
the stream entities of the integrity Preserve data
Conversational pattern stream integrity
(stringent and low
delay)
Examples of the voice, Streaming Web browsing, Background
application videotelephony multimedia network games download of emails
video games
One way communications is no communications

Location Services (LCS)
SMLC
UE
Node B
LMU
type B
HLR
Gateway
MLC
External
LCS client
LeLg
Lh
LMU
type A
Um
Iu
Iub
gsmSCF
Lc
MSC
BSC
BTS
LMU
type B
A/ (Gb)/
(Iu)
Abis
SRNC
SMLC
Lb
Ls
Uu
<- alternative ->
(R98 and 99)
<- alternative ->
SMLC
Lp
UTRAN
GERAN
Cell ID based
Observed Time Difference Arrival – Idle Period Downlink (OTDOA-IPDL)
Network Assisted GPS
You can run but you cannot hide

3G WCDMA and CDMA2000 Standards
UMTS-WCDMA CDMA2000
"No' Backward Compatibility Backward compatibility with CDMAOne
Cell Sites not synchronized Cell sites synchronized thru' GPS timing
Each cell site with different scrambling Adjacent cell sites use diffferent time offset
code for spreading of same scrambling code for spreading
Complex soft Hand Over Simple Soft Hand Over
Scrambling code 38,400 chips; frame Preudo Random (PN) sequence of length
of 10 ms 2
15
- 1 chips; period of 26.67 ms; different
site offset of 64 chips
OVSF Codes Walsh Codes

CDMA 2000 Layered Structure
Unique to cdma2000
Signaling
Services
Packet Data
Application
Packet Data
Application
Packet Data
Application
TCP UDP
IP
PPP
High Speed
Circuit Network
Layer Services
LAC Protocol Null LACLAC
MAC
Control
State
Best Effort Delivery RLP
QoS ControlMultiplexing
MAC
Physical Layer
Upper
Layers
(OSI 3-7)
Link
Layer
(OSI 2)
Physical
layer
(OSI 1)

UMTS Spectrum Allocation
Europe
Japan
Korea
USA
1800 1850 1900 1950 2000 2050 2100 2150 2200
GSM 1800
DL DECT
IMT-2000
TDD
IMT-2000
UL
MSS
UL
IMT-2000
TDD
IMT-2000
DL
MSS
DL
PHS
IMT-2000
UL
IMT-2000
DL
IMT-2000
DL
IS-95
DL
IMT-2000
UL
PCS/UL PCS/DL

WCDMA Circuit Switched Protocols
PHY
Phy-up
MAC
RLC
RRC
MM
CM
ATM
AAL2
FP
AAL5
SSCOP
SSCF-UNI
SSCOP
PHY
AAL5
SSCF-UNI
ALCAPNBAP
Phy-up
MAC
RLC
RRC
PHY
ATM
Q.2630.1
Q.2150.1
MTP3b
SSCF-NNI
SSCOP
AAL5
Iu
UP
AAL2
PHY
ATM
Q.2630.1
Q.2150.1
MTP3b
SSCF-NNI
SSCOP
AAL5
Iu
UP
AAL2
PHY
ATM
AAL2
FP
AAL5
SSCOP
SSCF-UNI
SSCOP
PHY
AAL5
SSCF-UNI
ALCAP NBAP
UE Node B RNC Core
RANAP
AAL5
SSCOP
SSCF-NNI
SCCP
MTP3B
RANAP
AAL5
SSCOP
SSCF-NNI
MM
CM
SCCP
MTP3B
CODEC

WCDMA PACKET CONTROL PLANE PROTOCOLS
SM
GMM
RRC
RLC
MAC-cd
PHY-up
FP FP
PHY-up
MAC-cd
RLC
RRC
NBAPALCAP ALCAPNBAP
AAL2
SSSAR
AAL2
SSSARSAALSAAL SAALSAAL
AAL5AAL5 AAL5AAL5
ATM ATM
PHY PHY
PHY
CDMA
PHY
CDMA
UE/MTE NODE B RNC SGSN
Uu Iub Iu-ps

WCDMA PACKET USER PLANE PROTOCOLS
IP
RLC
PDCP
MAC
PHY-up
FPALCAP
PHY-up
MAC
RLC
PDCP
AAL2SAALAAL2 SAAL
FP ALCAP
ATM ATM
PHY PHY
PHY
CDMA
PHY
CDMA
UE/MTE NODE B RNC SGSN
Uu Iub Iu-ps

HSDPA Protocol Architecture
L2
L1
HS-
DSCH
FP
RLC
L2
L1
L2
L1
L2
L1
HS-
DSCH
FP
Iub Iur
PHY
MAC
PHY
RLC
Uu
MAC-
hs
HS-
DSCH
FPHS-
DSCH
FP
MAC-c/sh
MAC-D

IMS Architecture
UTRAN
Home
Serving PS domain
IMS
Home
Serving PS domain
IMS
S-CSCF
I-CSCF
GGSNSGSN
HSS
P-CSCF
Other IP/IMS
network

Standards
 IEEE 802.11a and b: Wireless LAN (WiFi)
 IEEE 802.15: Wireless PAN (Bluetooth)
 IEEE 802.16d and e: Wireless MAN (WiMAX)
 IS-41: Inter-systems operation (TIA/EIA-41)
 IS-54: 1st
Gen (US) TDMA; 6 users per 30 KHz channel
 IS-88: CDMA
 IS-91: Analog Callular air interface
 IS-93: Wireless to PSTN Interface
 IS-95: TIA for CDMA (US) (Cdmaone)
 IS-124: Call detail and billing record
 IS-136: 2nd
Genr TDMA (TDMA control channel)
 IS-637: CDMA Short Message Service (SMS)
 IS-756: TIA for Wireless Network Portability (WNP)
 IS-2000: cdma2000 air interface (follow on to TIA/EIA 95-B)

R-SGW
Gi
Mr
Gi
Ms
MGW
MGCF
MRF
PSTN/
Legacy/Externa
l
Mm
Mw
Legacy
mobile
signaling
Network
Mc
Cx
Alternative
Access
Network Mh
CSCF
Mg
T-SGW
CSCF
HSS
MSC Server
Gi
MGW
GMSC Server
Nb
Mc Mc
Nc
T-SGW
Iu
3G All-IP Reference Architecture
Iu
Gi
R Uu
Gn
Gc
Gp
Signalling and Data Transfer Interface
Signalling Interface
Gr
Other PLMN
Gn
Applications
& Services
SCP
CAP
TE MT
SGSN
GGSN
HLR
SGSN
GGSN
Multimedia
IP Networks
UTRAN

N_B
PSTN/ISDN
N_B
RNC
RNC
Iub
IubIur
Internet/Intranet/ISP
Application
servers
Wireles
s
Data
Server
www,
email
IP
SGSN
GGSN
IP
Firewall
HLR
AuC
PCM
SS7
3G-MSC
ATM GTP+/IP
N_B
Internet/Intranet/ISPPSTN/ISDN
Application
servers
N_B
RNC
RNC
Iub
IubIur
ATM GTP+/IP
Wireless
Data
Server
www,
email
IP
PCM
SS7
IP
Firewall
GGSN
IP
PSTN/ISDN
MGCFHSS CSCF SGW
MGW
MRF
(G)MSC
Server
MGW3G-MSC
SGSN
GGSN
WCDMA 3G Evolution to All-IP Network
UTRAN

3.5G Radio Network Evolution
 High Data rate, low latency, packet optimized radio access
 Support flexible bandwidth up to 20 MHz, new transmission schemes, advanced
multi-antenna technologies, and signaling optimization
 Instantaneous peak DL 100 Mb/s and UP 50 Mb/S within 20 MHz spectrum
 Control plane latency of < 100 ms (camped to active) and < 50 ms (dormant to
active)
 > 200 users per cell within 5 MHz spectrum
 Spectrum flexibility from 1.25 MHz to 20 MHz
 Eliminate “dedicated” channels; avoid macro diversity in DL
 Migrate towards OFDM in DL and SC-FDMA in UL
 Support voice services in the packet domain
 Adaptive Modulation and Coding using Channel Quality Indicator (CQI)
measurements

3.5G WCDMA Evolved System Architecture
Evolved Packet Core
Evolved RAN
S1 Gi
Op.
IP
Serv.
(IMS,
PSS,
etc…)
Rx+
S2
GERAN
UTRAN
GPRS Core
Gb
Iu
S3
MME
UPE
Inter AS
Anchor
S4
non 3GPP
IP Access
HSS
PCRF
S5
S2
S7
S6
WLAN
3GPP IP Access
* Color coding: red indicates new functional element / interface
Source: www.3gpp.org

Upcoming
3.5 G
 Evolved radio Interface
 IP based core network
4G
 New Air Interface
 Very high bit rate services
 Convergence of Wireline, Wireless, and IP
worlds
And Now for Something Completely Different

Why Move Towards 4G?
 Limitation to meet expectations of applications like multimedia, full motion
video, wireless teleconferencing
Wider Bandwidth
 Difficult to move and interoperate due to different standards hampering global
mobility and service portability
 Primarily Cellular (WAN) with distinct LANs’; need a new integrated network
 Limitations in applying recent advances in spectrally more efficient modulation
schemes
 Need all digital network to fully utilize IP and converged video and data
Incessant human desire to reach the sky

Where Do We Want to Go?
 Seamless Roaming
 Integrated “standard” Networks
 Mobile Intelligent Internet
 Onwards to (Ultra) Wideband Wireless IP Networks
We are no longer in Kansas, Toto

 It is a framework to meet the need of a universal highspeed wireless
networks.
 It supports Interact multimedia services such as Tele conferrencing wireless
Internet over wide bandwidth with higher data rate.
 It will will be in a reasonable low cost than previous Generation.
 Still in the cloud of ITU and IEEE of 3GPP LTE from UMTS and WI -MAX
4’th Generation

New in 4G
Entirely Packet Switched Network
All Networks are Digital
Higher bandwidth at Low cost
(up to 100 mbps)
Tight Network Security
Potential Application :
Virtual Presence
Virtual Navigation
Tele Medicine
Tele geo-Processing
Crisis- management application
Education purpose

 Mobile IP
VoIP
Ability to move around with the same IP address
IP tunnels
Intelligent Internet
 Presence Awareness Technology
Knowing who is on line and where
 Radio Router
Bringing IP to the base station
 Smart Antennas
Unique spatial metric for each transmission
Wireless IP <---> IP Wireless

4G Networks Advances
 Seamless mobility (roaming)
—Roam freely from one standard to another
—Integrate different modes of wireless communications – indoor
networks (e.g., wireless LANs and Bluetooth); cellular signals; radio and TV;
satellite communications
 100 Mb/se full mobility (wide area); 1 Gbit/s low mobility (local area)
 IP-based communications systems for integrated voice, data, and video
—IP RAN
 Open unified standards
 Stream Control Transmission Protocol (SCTP)
—Successor to “SS7”; replacement for TCP
—Maintain several data streams within a single connection
 Service Location Protocol (SLP)
—Automatic resource discovery
—Make all networked resources dynamically configurable through IP-based
service and directory agents
The demise of SS7

802.11a/g
1995 2000
200 Mbps
~ 14.4 kbps
50 Mbps
144 kbps
2010+
384 kbps
2005
2G
(Digital)
1G
(Analog)
2.4 GHz
WLAN
802.11b
4G
PAN
5 GHz
WLAN
3G
(IMT2000)
Fast
Slow
1Gbps
OFDM
A
CDMA
5 GHz
WLAN
802.11n
WiBro

Future Enhancement
• By Using 4G Technology, we aimed to
prepare that “Open Smart Classroom” in
Real time application using
Wi-Bro
with
• Specs with virtual Screen .
• Multi Language Translation.
WiBro (Wireless + Broadband)

Key 3G and 4G Parameters
Attribute 3G 4G
Major Characteristic Predominantly voice- data as
add-on
Converged data and VoIP
Network Architecture Wide area Cell based Hybrid – integration of
Wireless Lan (WiFi), Blue
Tooth, Wide Area
Frequency Band 1.6 - 2.5 GHz 2 – 8 GHz
Component Design Optimized antenna; multi-
band adapters
Smart antennas; SW multi-
band; wideband radios
Bandwidth 5 – 20 MHz 100+ MHz
Data Rate 385 Kbps - 2 Mbps 20 – 100 Mbps
Access WCDMA/CDMA2000 MC-CDMA or OFDM
Forward Error Correction Convolution code 1/2, 1/3;
turbo
Concatenated Coding
Switching Circuit/Packet Packet
Mobile top Speed 200 kmph 200 kmph
IP Multiple versions All IP (IPv6.0)
Operational ~2003 ~2010

The development of the mobile communication system
LTE
3G
2G
1G
Using the cellular network, widely used
standards AMPS, TACS, etc., using analog
technology and frequency division multiple
access (FDMA) technology.
The most widely used communication system,
including GSM, IS-95,etc., digital technology, Using
FDM, TDM, CDMA technology. Providing digitized
voice services and low-speed data services
International standards includes WCDMA, CDMA2000, TD-SCDMA, WiMax.
Technical indicators: Indoor rate is 2Mbps, outdoor rate is 384kbps, traffic
rate is 144kbps. Providing voice services, high-speed transmission ,
broadband multimedia services, wireless access to the Internet and so on.
OFDM and MIMO technology , in the 200MHz system
bandwidth, peak rate of downlink is 100Mbps, peak
rate of uplink is 50MHz. Providing high-rate data
transmission services such as VoIP and IMS.

UMTS long-term evolution
——The 3.9G era of LTE
The third
generation of
mobile
communication
technology
HSPA
Evolution to
LTE
802.16m
Wimax
technology
To evolution
along EV-DO
Rev.0/Rev.A/Rev.
B to UMB

The essence of LTE is the
contradictions and
unification between the
IEEE implemented
broadband access mobile
and 3GPP pursues
broadband mobile
communications.

LTE deployment in China
The first TD-LTE
demonstration network
in Shanghai World Expo
Xiamen: 100 LTE
base station
Guangzhou Asian
Games: TD-LTE
trial network
Zhuhai: 100 LTE base
station

Higher
(higher data rates, higher
spectral efficiency)
Faster
(low delay)
Stronger
(based on full-packet
and Large throughput)

Higher performance, lower cost
Throughput Delay 1M byte costs Mobility
roaming

Two Frame structure :FDD and TDD
Frame structure 1——Apply to FDD
#0 #1 #2 #3 #18 #19
One Subframe
A radio frame which is suitable for FDD 1 frame structure is 10ms,
contains10 sub-frames, each sub-frame is 1ms, including two slots,
each slot is 0.5ms.
One Radio frame,
One Time slot,

DwPTS
GP UpPTS DwPTS GP UpPTS
Subframe 0 Subframe 2 Subframe 3
Frame structure 2——Apply to TDD
One Radio frame,
A Half-frame,
One Time
slot
A subframe
Subframe 4 Subframe 5 Subframe 7 Subframe 8 Subframe 9

OFDM
(Orthogonal Frequency
Division Multiplexing)

 A basic requirement of the future mobile communication system is the
high data rate, but the high-speed data transmission of a communication
system is often subject to ISI and frequency selective fading caused by
multipath interference.
 This phenomenon seems that one - way street road often will result rear-
end collision inter-Vehicle (inter-symbol interference of the vehicle),
because of the vehicle excessive ,in order to prevent the generation of the
rear-end, thereby to reduce the speed through extension into multiple
carriageway .
 In LTE, in order to combat the inter-symbol interference and frequency
selective fading in multipath channel , we adopt narrow-band parallel data
transmission with cycle prefix , which transforms high-speed data flow to
multiplexed parallel data low-speed flow, namely, this transmission mode
is the OFDM.

……
…
……
…
MIMO+OFDM

MIMO can be roughly divided into three kinds :
transmission diversity, the spatial multiplexing and beamforming
 Transmission diversity ： providing more data flow copy by use of the weak
correlation of large space antenna or beam space between the channel ， so
as to improve the reliability of the channel and to reduce the bit error rate.
 Space reuse ： the process which make use of the weak correlation of large
antenna spacing between the channel to transfer different data flow in the
corresponding channel . It is worth noting that the transmission diversity is the
transmission of the same data flow in different channel, while spatial
multiplexing is the transmission different data streams in different channel.
 Beamforming : achieved by directivity of flowing antenna and leting the
electromagnetic wave from the antenna coming towards the direction of users.

LTE-Advanced
Peak Rate
Under the condition of low speed, IMT-Advancedte technology demands the
peak rate at a rate of 1 Gbps
Under the condition of high speed ， IMT-Advanced technology demands the
peak rate at a rate of 100 Gbps
The peak rate of uplink reaches 1 Gbps
The peak rate of downlink reaches 500 Mbps
Time Delay
resident status
less than 50ms
activation
（ in-
sync ）
activation -“dormancy ”
（ un-sync ）
less than
10ms
Demand

LTE-Advanced technological evolution
20MHz
20MHz
100MHz
60MHz
Carrier Polymerization——Integration Of Resources

• Joint Processing
eNod
eB
eNod
eB
eNod
eB
RRU
UE
TP(Serving
cell)
TP
Multipoint Transmission Joint Transmission
COMP——Many hands make light work

• Cooperation Scheduling/Coordinated Beamforming （ CS/CB ）
TP1 TP2
UE1
UE2
UE3
COMP——Many hands make light work

Self-organized Network——Keep your business to yourself

eNB power on
（ or cable
connection ）
(A) Basic start
(B) Initialization
infinite configuration
(C) Optimization
/ Self-adaption
a-1:IP address configuration &operation
and maintenance system
a-2:Authentication of eNB/NW
a-3:connection to aGW
a-4:Dowmload eNB (and operating
parameters)
b-1:Adjacent village list configuration
b-2:Correlation parameter
configuration of covering capacity
c-1:Adjacent village list optimization
c-2:Coverage and capacity control
Self-Configuration
（ Preliminary running
state ）
Self-Optimizing
（ Running
state ）
Family base station——I’m not WIFI

Network access securityNetwork domain safetyUser domain safety
Application domain safetyVisualization and
Configuring security
Information Security of B3G and 4G
Customer
Application
Supplier's Application
USI
M
Mobile
Equipment AN
Service
Network
HE
Transmission
layer
Local layer
/Service
layer
Application
layer
（Ⅳ）
( )Ⅰ ( )Ⅰ
( )Ⅰ
( )Ⅰ
( )Ⅱ
( )Ⅱ
( )Ⅰ( )Ⅰ(Ⅲ
)
System Security Framework
Network Domain Security

The security demand of LTE
 The user to network security
• User identity and Device security
• User data and Signalling safety
USIM
Card
Encrypt ？
Valid User
Should I encrypt between
mobile station and USIM?
 Safety visibility and Configurability

 Base station (eNB) safety
S
1
S
1
X2
I must be legal,
and you?
Are you legal ?
Authentication Between Base Stationes

Cellular systems and infrastructure base wireless network

More Related Content

What's hot (20)

Similar to Cellular systems and infrastructure base wireless network (20)

Recently uploaded (20)

Cellular systems and infrastructure base wireless network

Editor's Notes