UNIT-1

Wireless propagation mechanisms, Statistical Multipath

Channel Models and evolution from 1G to 4G
Channel Models: INTRODUCTION
• For the design, simulation, and planning of wireless systems, we need models for the propagation
channels.
• There are two main applications for channel models:
1. For the design, testing, and type approval of wireless systems, we need simple channel models
that reflect the important properties of propagation channels – i.e., properties that have an impact
on system performance. This is usually achieved by simplified channel models that describe the
statistics of the impulse response in parametric form. The number of parameters is small and
independent of specific locations. Such models sometimes lead to insights due to closed-form
relationships between channel parameters and system performance. Furthermore, they can easily
be implemented by system designers for testing purposes.

2. The designers of wireless networks are interested in optimizing a given system in a certain
geographical region. Locations of Base Stations (BSs) and other network design parameters
should be optimized on the computer, and not by field tests, and trial and error. For such
applications, location-specific channel models that make good use of available geographical and
morphological information are desirable. However, the models should be robust with respect to
small errors in geographical databases.
• The following three modelling methods are in use for these applications:
1. Stored channel impulse responses:
A channel sounder measures, digitizes, and stores impulse responses h(t, τ ).
The main advantage of this approach is that the resulting impulse responses
are realistic. Furthermore, system simulations using the stored data are
reproducible, as the data remain available and can be reused indefinitely, even
for simulations of different systems. This is an important distinction from field
trials of whole systems, where there can be no guarantee that the impulse
response remains constant over time. The disadvantages of using stored
impulse responses are (i) the large effort in acquiring and storing the data and
(ii) the fact that the data characterize only a certain area, and need not be
typical for a propagation environment.
An example of stored CIR readings can be found below:
2. Deterministic channel models:

These models use the geographical and morphological information from a

database for a deterministic solution of some approximation thereof. The
basic philosophy is the same as for stored impulse responses: determining
the impulse response in a certain geographic location. Both of these
methods are therefore often subsumed as site-specific models. The
drawbacks of deterministic (computed) channel models compared with
stored (measured) impulse responses are (i) the large computational effort
and (ii) the fact that the results are inherently less accurate, due to
inaccuracies in the underlying databases and the approximate nature of
numerical computation methods. The main advantage is that computer
simulations are easier to perform than measurement campaigns.
3. Stochastic channel models:
These models use the probability density function (pdf) of the channel impulse response (or equivalent functions).
These methods do not attempt to correctly predict the impulse response in one specific location, but rather to
predict the pdf over a large area. The simplest example of this approach is the Rayleigh-fading model: it does not
attempt to correctly predict the field strength at each location, but rather attempts to correctly describe the pdf of
the field strength over a large area. Stochastic wideband models can be created in the same spirit.
 Narrowband Models: Modelling of Small-Scale and Large-Scale Fading
• For a narrowband channel, the impulse response is a delta function with a time-varying
attenuation, so that for slowly time-varying channels: h(t, τ ) = α(t)δ(τ ).

• The variations in amplitude over a small area are typically modelled as a random
process, with an autocorrelation function that is determined by the Doppler spectrum.
The complex amplitude is modelled as a zero-mean, circularly symmetric complex
Gaussian random variable. As this gives rise to a Rayleigh distribution of the absolute
amplitude, we henceforth refer to this case simply as “Rayleigh fading.”

• When considering variations in a somewhat larger area, the small-scale averaged

amplitude F obeys a lognormal distribution, with standard deviation σF; typically,
values of σF are 4 to 10 dB. The spatial autocorrelation function of lognormal
shadowing is usually assumed to be a double-sided exponential, with correlation
distances between 5 and 100 m, depending on the environment.
 Narrowband models : Pathloss model
OKUMURA MODEL
 It is one of the most widely used models for signal prediction in urban areas, and it is applicable for
frequencies in the range 150 MHz to 1920 MHz
 Based totally on measurements (not analytical calculations)
 Applicable in the range: 150MHz to ~ 2000MHz, 1km to 100km T-R separation, Antenna heights of
30m to 100m
 Okumura Model
 The major disadvantage with the model is its low response to rapid changes in terrain, therefore the
model is fairly good in urban areas, but not as good in rural areas.
 Common standard deviations between predicted and measured path loss values are around 10 to 14 dB.
 G(hre)：

 hte 
G (hte )  20 log   1000m  hte  30 m
 200 

 hre 
G (hre )  10 log   hre  3 m
 3 

 hre 
G (hre )  20 log   10m  hre  3 m
 3 
 Okumura and Hata’s model
 Example 4.10
 Hata Model
 Empirical formulation of the graphical data in the Okamura model. Valid 150MHz to
1500MHz, Used for cellular systems
 The following classification was used by Hata:
■Urban area LdB  A  B log d  E
■Suburban area LdB  A  B log d  C
■Open area LdB  A  B log d  D
A  69.55  26.16 log f  13.82hb
B  44.9  6.55 log hb
C  2(log( f / 28)) 2  5.4
D  4.78 log( f / 28) 2  18.33 log f  40.94
E  3.2(log( 11.75hm )) 2  4.97 for large cities, f  300MHz
E  8.29(log( 1.54hm )) 2  1.1 for large cities, f  300MHz
E  (1.11log f  0.7)hm  (1.56 log f  0.8) for medium to small cities
Statistical Multipath Channel Models

• Random # of multipath components, each with

 Random amplitude
 Random phase
 Random Doppler shift
 Random delay
• Random components change with time
• Leads to time-varying channel impulse response
Time-Varying Channel Impulse Response
• Let the transmitted signal be as:
𝑠 𝑡 = 𝑅𝑒{𝑢(𝑡)𝑒 𝑗2𝜋𝑓𝑐 𝑡 } = 𝑅𝑒 𝑢 𝑡 𝑐𝑜𝑠2𝜋𝑓𝑐 t −𝐼𝑚 𝑢 𝑡 𝑠𝑖𝑛2𝜋𝑓𝑐 t
where u(t) is the equivalent low-pass signal for s(t) with bandwidth Bu and where fc
is its carrier frequency. Neglecting noise, the corresponding received signal is the
sum of the line-of-sight path and all resolvable multipath component:
𝑁(𝑡)

𝑟 𝑡 = 𝑅𝑒 ෍ 𝛼𝑛 𝑡 𝑢(𝑡 − 𝜏𝑛 𝑡 )𝑒 𝑗(2𝜋𝑓𝑐 𝑡−𝜏𝑛 𝑡 +𝜑𝐷𝑛 )

𝑛=0
where n = 0 corresponds to the LOS path. The unknowns in this expression are: the
number of resolvable multipath components N(t); and, for the LOS path and each
multipath component, its path length 𝑟𝑛 (t) and corresponding delay 𝜏𝑛 (t) =𝑟𝑛 (t)/c
where “c” is speed of light, Doppler phase shift 𝜑𝐷𝑛 (t), and amplitude α𝑛 (t).
 • The nth resolvable
multipath component
may result from a single
reflector or with multiple
reflectors

• Single reflector
•  n (t) is a function of the single reflector
 j 2  f c  n  
•  n t   e is the phase shift
v cos  n  t 
• fD  is the Doppler shift
n

• D n
  2 f t  d t
Dn
is the Doppler phase shift
t

16
 • Reflector cluster
• Two multipath components
with delay  1 and  2 are
"resolvable" if their delay
difference considerably
exceeds the inverse signal
bandwidth
1
1 2   1
Bu

• If u (t -  1) ~ u (t -  2) the two components cannot

be separated at the receiver and are "unresolvable"
• Unresolvable signals are usually combined into a
single term with delay 
17
• The amplitude of unresolvable signals will typically
undergo fast variations due to the constructive and
destructive combining
• Typically, wideband channels will be resolvable
while narrowband channels may not
• Since  n (t),  n (t), and  Dn (t) change with time
they are characterized as random processes
• The received signal is also stationary and ergodic
(can be characterized from a sample)

18
• We say that two multipath components with delay 𝜏1 and 𝜏2 are
resolvable if their delay difference significantly exceeds the inverse signal
bandwidth:|𝜏1 − 𝜏2 | >> 𝐵𝑢−1 . Multipath components that do not satisfy
this resolvability criteria cannot be separated out at the receiver because
u(t − 𝜏1 ) ≈ u(t − 𝜏2 ), and thus these components are non resolvable.
These non resolvable components are combined into a single multipath
component with delay τ ≈ 𝜏1 ≈ 𝜏2 and an amplitude and phase
corresponding to the sum of the different components. The amplitude of
this summed signal will typically undergo fast variations due to the
constructive and destructive combining of the non resolvable multipath
components. In general, wideband channels have resolvable multipath
components so that each term in the summation of (eqn 2) corresponds
to a single reflection or multiple non resolvable components combined
together, whereas narrowband channels tend to have non resolvable
multipath components contributing to each term in (eqn 2).
Time Varying Impulse Response:
• Response of channel at t to impulse at t-τ:
𝑁(𝑡)
𝑐(τ, t) = σ𝑛=0 𝛼𝑛 (t).𝑒 −𝑗∅𝑛 (𝑡) .𝛿(𝜏 − 𝜏𝑛 (t))

Fig: System multipath at two different measurement times.

• t is time when impulse response is observed
• t-τ is time when impulse put into the channel
• τ is how long ago impulse was put into the channel for the current observation
• path delay for MP component currently observed
Received Signal Characteristics:
• Received signal consists of many multipath components
• Amplitudes change slowly
• Phases change rapidly
 Constructive and destructive addition of signal components
 Amplitude fading of received signal (both wideband and narrowband signals)
Narrowband Fading Models:
Narrowband Fading Approximation:
• Assume delay spread 𝑴𝒂𝒙𝒎.𝒏 𝝉𝒏 𝒕 − 𝝉𝒎 (𝒕) ≪ 𝟏/𝑩
• Then u(t) is approximately equal to u(t-τ).
𝑵(𝒕)
• Received signal given by: 𝒓 𝒕 = 𝑹 𝒖(𝒕)𝒆𝒋𝟐𝝅𝒇𝒄 𝒕 σ𝒏=𝟎 𝜶𝒏 𝒕 . 𝒆−𝒋𝝋𝒏 (𝒕)
• The scale factor is independent of the transmitted signal s(t) and, in particular, of the equivalent low-pass signal
u(t) as long as the narrowband assumption Tm<<1/B is satisfied.
• With these above assumption the received signal becomes:
𝑵(𝒕)
𝒓 𝒕 = 𝑹 𝒆𝒋𝟐𝝅𝒇𝒄𝒕 σ𝒏=𝟎 𝜶𝒏 𝒕 . 𝒆−𝒋𝝋𝒏 (𝒕) = 𝒓𝑰 𝒕 𝒄𝒐𝒔𝟐𝝅𝒇𝒄 𝑡 − 𝑟𝑄 𝑡 𝑠𝑖𝑛𝟐𝝅𝒇𝒄 𝑡
where the in-phase and quadrature components are given by

and where the phase term is now incorporates the phase

offset 𝜑𝑜 as well as the effects of delay and Doppler.
 Cross - Correlation
• In signal processing, the cross-correlation is a
measure of similarity of two signals
• Used to find features in an unknown signal by
comparing it to a known one
• It is a function of the relative time between the
signals

23
For discrete functions f i and g i the cross-
correlation is defined as

f  g  i  f
j
j * gi  j

For continuous functions f (x) and g i the

cross-correlation is defined as

 f  g  x    f *  t  g  x  t  d t

24
 Properties of Cross - Correlation
• Similar in nature to the convolution of two functions
• They are related by

f t   g t   f *   t   g t 
if f (t) or g (t) is an even function

25
 Autocorrelation
• Autocorrelation is the cross-correlation of a signal with
itself
• Autocorrelation is useful for finding repeating patterns
in a signal
• Determining the presence of a periodic signal which
has been buried under noise
• Identifying the fundamental frequency of a signal
which doesn't actually contain that frequency
component, but implies it with many harmonic
frequencies
• Different definitions in statistics and signal processing

26
 Signal processing

• Given a signal f(t), the continuous autocorrelation

R f () is the continuous cross-correlation of f (t)
with itself, at lag , and is defined as
 
R f    f *    f     f t    f * t  d t   f t  f * t    d t
 

Basically, autocorrelation is the convolution of a

signal with itself

Note that, for a real function, f (t) = f * (t)

27
Formally, the discrete autocorrelation R at lag j
for signal x n is

R j   x
n
n m  x n j m 
For zero – centered signals (zero mean)

R j  x
n
n x n j

28
• A fundamental property of the autocorrelation
function is symmetry, R(i) = R(− i)
• In the continuous case, R f (t) is an even function
R f     R f   when f (t) is real
R f     R f *   when f (t) is complex

• The continuous autocorrelation function

reaches its peak at the origin, where it takes a
real value
R f    R f  0 
The same result holds in the discrete
case

29
• The autocorrelation of a periodic function is, itself,
periodic with the very same period
• The autocorrelation of the sum of two completely
uncorrelated functions (the cross-correlation is zero
for all τ) is the sum of the autocorrelations of each
function separately
• Since autocorrelation is a specific type of cross-
correlation, it maintains all the properties of cross-
correlation
• The autocorrelation of a white noise signal will have a
strong peak at  = 0 and will be close to 0 for all other 
• A sampled instance of a white noise signal is not
statistically correlated to a sample instance of the
same white noise signal at another time
30
 Autocorrelation, Cross – Correlation, and Power
Spectral Density for Narrowband Model
• Assumptions:
• No dominant LOS component
• Each of the multipath components is associated
with a single reflector
•  n (t)   n = constant
•  n (t)   n = constant
• f Dn (t)  f Dn = constant
•  Dn (t)  2  f Dn t
•  n (t)  2  f c  n + 2  f Dn t -  0
• 2  f c  n changes more rapidly than the others
•  n (t) is uniformly distributed on [- , ] 31
 Under these Assumptions:
 
E r I  t   E   n cos  n  t  
   E  n
 E cos  n  t   0
n  n

Similarly, E r Q  t    0

Therefore, E r  t    0

Zero – mean Gaussian process

If there is a dominant LOS product, the assumption of a

random uniform phase no longer holds

32
Correlation of In – Phase and Quadrature Components

By the same process: E r I  t  r Q  t    0

• Conclusions:
• r I (t) and r Q (t) are uncorrelated
• They are independent

33
 Autocorrelation of In – Phase Component

Ar I  t , t     E r I  t  ,r I t   

We can show that this expression is equal to

n
 
A r I  t , t     0.5  E  n2  E cos 2  f D n  
 
   n 
 0.5  E   E cos  2    cos  
2

n 
n
  
n
f D   cos  constant
n

A r  t , t     A r  
I I
and A r  t , t     A r  
Q Q

Where this is the case, we say that r I (t) and r Q (t) are
wide – sense stationary (WSS) random processes
34
 Cross – Correlate A r I  t , t    and A r Q  t , t   

A r I ,r Q  t , t     A r I ,r Q    E r I  t  r Q  t    
 n 
  0.5  E   sin  2    cos 
2

n
n
  
  E r Q  t  r I  t    

The received signal

  
r  t   r I  t  cos 2  f c t  r Q  t  sin 2  f c t 
is also WSS with autocorrelation

  
Ar    E r  t  r t     A r I   cos 2 f c   A r Q   sin 2 f c  
35
 Uniform Scattering Environment
Many scatterers densely packed wrt angle

Dense Scattering
Environment

36
Assumptions:
• N multipath components with angle of arrival

 n  n 
2 2
   N
N 
2 Pr
E   
2
n
N
• P r = Total received power

Pr N
 n  
Ar I  
N n 1
cos  2   cos
  

37
 2
Substitute N 

Pr 
N
n  
Ar I  
2 n 1
cos  2   cos
   

Take the limit as N 

2
Pr  
Ar I  
2 0  

cos  2   cos  d   Pr J 0 2  f D  

1
J0 x   e  j x cos 
d
 0

J 0 (x) is a Bessel function of the zeroth order

38
 2
Pr  
Similarly, Ar Q  
2 0
sin  2   cos  d   0
 

Autocorrelation is
zero when f D   0.4

  = 0.4 

Independent at
this point
Recorrelate later
The expression   = 0.4  turns out to be very significant
and dictates such actions as antenna spacing 39
 Power Spectral Densities
• Take Fourier Transforms of the autocorrelation
functions of r I (t) and r Q (t)
2 Pr 1
S r I  f   S r Q  f     A r I     f  fD
 fD  f 
2

1  
 fD
 
0 elsewhere

40
 Recall:
Ar    E r  t  r t     A r   cos  2 f c    A r   sin  2 f c  
I Q

PSD of the received signal r (t) under uniform scattering

  
S r  f     A r     0.25 S r I f  f c  S r I f  f c 
Pr 1
 f  fc  fD
2 f D  f  fc 
2

1  
 fD 
 
 0 elsewhere

This expression integrates to P r as required

41
Wideband Fading Models: Power Delay Profile, Coherence Bandwidth, Doppler Power
Spectrum and Channel Coherence Time.
• When the signal is not narrowband we get another form of distortion due to the multipath delay spread. In this
case a short transmitted pulse of duration T will result in a received signal that is of duration T +Tm, where Tm is
the multipath delay spread. Thus, the duration of the received signal may be significantly increased. This
phenomenon is illustrated in Figure below. In the figure, a pulse of width T is transmitted over a multipath
channel.

• Linear modulation consists of a train of pulses where each pulse carries information in its amplitude and/or phase
corresponding to a data bit or symbol. If the multipath delay spread Tm<<T then the multipath components are
received roughly on top of one another, as shown in the upper right of the figure. The resulting constructive and
destructive interference causes narrowband fading of the pulse, but there is little time spreading of the pulse and
therefore little interference with a subsequently transmitted pulse. On the other hand, if the multipath delay
spread Tm>>T, then each of the different multipath components can be resolved, as shown in the lower right of
the figure. However, these multipath components interfere with subsequently transmitted pulses (dashed pulses
in the figure). This effect is called inter-symbol interference (ISI).
Power Delay Profile:
• Power delay profile P ( τ ): the channel power spectral density as a function of delay, i.e.
how “channel power” is distributed along dimension excess delay τ.

• Consider a local area around a spatial position, averaging ℎ𝐵 (𝑡, 𝜏) 2 over time gives
rise to P(τ).
• Specifically, P(τ) is Fourier transform of autocorrelation function of ℎ𝐵 (𝑡, 𝜏)
• Mean excess delay is defined as the first moment of power delay profile:
• Root mean square (RMS) delay spread is defined as the square root of the second central moment of
power delay profile:

where the second moment is given by:

• Coherence bandwidth is a measure of the range of frequencies over which the channel is “flat” (i.e.
passing spectral components with approximately equal gain and linear phase)
1
𝐵𝑐 = ; where 𝑇𝑟𝑚𝑠 =rms delay spread (also denoted by 𝜎𝜏 ).
2𝜋𝑇𝑟𝑚𝑠
• Coherence time is the time duration over which the channel impulse response is considered to be not
varying. Such channel variation is much more significant in wireless communications systems, due to
Doppler effects.
where is Coherence Time and is the maximum Doppler spread or, maximum Doppler
frequency or, maximum Doppler shift given by where is the carrier frequency.
• If a sinusoidal signal is transmitted (represented by a spectral line in the frequency domain), after
transmission over a fading channel, we will receive a power spectrum that is spread according to the
image below. The frequency range where the power spectrum is nonzero defines the Doppler spread and
the power spectrum is known as Doppler Power Spectrum.
A Cellular Network
Cell 1

Mobile Public
Telephone Switched
Switching Telephone
Cell 2 Center Network
(MTSC) (PSTN)

HLR VLR

Mobile User Base Transceiver Station (BTS)

Cordless connection HLR = Home Location Register

Wired connection VLR = Visitor Location Register

Overview of Location Services
 Cell-id based location.
 assigned an id of the cell that you are in.
 cell-id is stored in a database.
 As you move from one cell to another, you are assigned a different cell-
id and the location database is updated.
 most commonly used in cellular networks. (HLR, VLR)
 Neighborhood polling: Connected mobile units only move to adjacent cells
 Angle of arrival (AOA). the angle at which radio waves from your device
"attack" an antenna is used to calculate the location of the device.
 Time taken. In this case, the time taken between the device and the antenna
is used to calculate the location of the device.
 Network assisted Global Positioning System (GPS). a GPS chip is installed
inside a phone and thus the location of the user is tracked.
 Cellular System
Cell 1

Mobile Public
Telephone Switched
Switching Telephone
Cell 2
Center Network
(MTSC) (PSTN)
HLR VLR

Handoffs (typically 30 mseconds):

1. At any time, mobile station (MS) is in one cell and under the control of a BS
2. When a MS leaves a cell, BS notices weak signal
3. BS asks surrounding BSs if they are getting a stronger signal
4. BS transfers ownership to one with strongest signal
5. MTSO assigns new channel to the MS and notifies MS of new boss
Frequency Reuse

The concept of frequency reuse is based on assigning to

each cell a group of radio channels used within a small
geographic area
Cells are assigned a group of channels that is completely
different from neighbouring cells
The coverage area of cells is called the footprint and is
limited by a boundary so that the same group of channels can
be used in cells that are far enough apart
Frequency Reuse
• Cells with the same
number have the
same set of
frequencies

Frequency Reuse
Frequency Reuse using 7 frequencies
allocations

f2
f7 f3 f2
f1 f7 f3
f6 f4 f1
f2 f5 f6 f4
f7 f3 f2 f5
f1 f7 f3 f2
f6 f4 f1 f7 f3
f5 f6 f4 f1
f5 f6 f4
f5

Each cell is generally 4 to 8 miles in diameter with a lower limit

around 2 miles.
Problem with Smaller Clustersize

Interfering cells are closer by when clustersize is smaller.

 0G Wireless
• Mobile radio telephones were used for military communications in
early 20th century
• Car-based telephones first introduced in mid 1940s
• Single large transmitter on top of a tall building
• Single channel used for sending and receiving
• To talk, user pushed a button, enabled transmission and disabled
reception
• Became known as “push-to-talk” in 1950s
• CB-radio, taxis, police cars use this technology
• IMTS (Improved Mobile Telephone System) introduced in 1960s
• Used two channels (one for sending, one for receiving)
• No need for push-to-talk
• Used 23 channels from 150 MHz to 450 MHz
 First-Generation Cellular
• Advanced Mobile Phone Service (AMPS) invented at Bell Labs and
first installed in 1982
• Used in England (called TACS) and Japan (called MCS-L1)
• Key ideas:
• Exclusively analog
• Geographical area divided into cells (typically 10-25km)
• Cells are small: Frequency reuse exploited in nearby (not adjacent) cells
• As compared to IMTS, could use 5 to 10 times more users in same area by
using frequency re-use (divide area into cells)
• Smaller cells also required less powerful, cheaper,smaller devices
Cellular Network Organization
• Cell design (around 10 mile radius)
• Served by base station consisting of transmitter, receiver,
and control unit
• Base station (BS) antenna is placed in high places (churches,
high rise buildings) -
• Operators pay around $500 per month for BS
• 10 to 50 frequencies assigned to each cell
• Cells set up such that antennas of all neighbors are
equidistant (hexagonal pattern)
• In North America, two 25-MHz bands allocated to
AMPS
• One for transmission from base to mobile unit
• One for transmission from mobile unit to base
Approaches to Increase Capacity
• Adding/reassigning channels - some channels are
not used
• Frequency borrowing – frequencies are taken from
adjacent cells by congested cells
• Cell splitting – cells in areas of high usage can be split
into smaller cells
• Microcells – antennas move to buildings, hills, and
lamp posts
Second Generation Cellular
• Based on digital transmission
• Different approaches in US and Europe
• US: divergence
• Only one player (AMPS) in 1G
• Became several players in 2G due to competition
• Survivors
• IS-54 and IS-135: backward compatible with AMPS frequency
allocation (dual mode - analog and digital)
• IS-95: uses spread spectrum
• Europe: Convergence
• 5 incompatible 1G systems (no clear winner)
• European PTT development of GSM (uses new frequency
and completely digital communication)
Advantages of Digital Communications for
Wireless
• Voice, data and fax can be integrated into a single system
• Better compression can lead to better channel utilization
• Error correction codes can be used for better quality
• Sophisticated encryption can be used
Differences Between First and Second
Generation Systems
• Digital traffic channels – first-generation systems are almost purely
analog; second-generation systems are digital
• Encryption – all second generation systems provide encryption to
prevent eavesdropping
• Error detection and correction – second-generation digital traffic
allows for detection and correction, giving clear voice reception
• Channel access – second-generation systems allow channels to be
dynamically shared by a number of users
Integrating Data Over Cellular
• Direct access to digital channel
• Voice and data using one handset
• PCS 1900 (GSM-1900)
• 9.6 kbps circuit switched data
• 14.4 kbps under definition
• Packet mode specified
• Short message service
• IS-95-based CDMA
• 13 kbps circuit switched data
• Packet mode specified
• Short message service
GSM (Global System for Mobile Communications)
• Completely designed from scratch (no backward compatability)
• Uses 124 channels per cell, each channel can support 8 users
through TDM (992 users max)
• Some channels used for control signals, etc
• Several flavors based on frequency:
• GSM (900 MHz)
• GSM 1800 (called DCS 1800)
• GSM 1900 (called DCS 1900) - used in North America
• GSM 1900 phone only works in North America.
• In Europe, you can transfer your SIM (Subscriber Identity Module)
card to a phone of the correct frequency. This is called SIM-
roaming.
GSM (2G-TDMA)
• Circuit mode data
• Transparent mode
• Non-transparent mode using radio link protocol
• Data rate up to 9.6kb/s
• Short message service
• Limited to 160 characters
• Packet mode data: Plans for GSM Phase 2+
• Architecture specification very detailed (500 pages)
• Defines several interfaces for multiple suppliers
Mobile Station and Base Station Subsystem (BSS)
Mobile station
• Mobile station communicates across Um interface (air interface) with
base station transceiver in same cell as mobile unit
• Mobile equipment (ME) – physical terminal, such as a telephone or PCS
• ME includes radio transceiver, digital signal processors and subscriber identity
module (SIM)
• GSM subscriber units are generic until SIM is inserted
• SIMs roam, not necessarily the subscriber devices
BSS
• BSS consists of base station controller and one or more base
transceiver stations (BTS)
• BSC reserves radio frequencies, manages handoff of mobile unit from
one cell to another within BSS, and controls paging
Network Subsystem Center
Mobile Switching Center (MSC) is at core; consists of
several databases
• Home location register (HLR) database – stores
information about each subscriber that belongs to it
• Visitor location register (VLR) database – maintains
information about subscribers currently physically in
the region
• Authentication center database (AuC) – used for
authentication activities, holds encryption keys
• Equipment identity register database (EIR) – keeps
track of the type of equipment that exists at the
mobile station
 GSM Location Services
6 2
9
BTS Terminating Gateway
10 MSC MTSC Public 1
10 10 Switched
9 10 7 8 5 3 Telephone
4 Network
(PSTN) 10
VLR HLR
5
6. Call routed to terminating MSC
1. Call made to mobile unit (cellular phone)
7. MSC asks VLR to correlate call to
2. Telephone network recognizes number
the subscriber
and gives to gateway MSC
8. VLR complies
3. MSC can’t route further, interrogates
9. Mobile unit is paged
user’s HLR
10. Mobile unit responds, MSCs convey
4. Interrogates VLR currently serving user
(roaming number request) information back to telephone
5. Routing number returned to HLR and
then to gateway MSC
Legend: MTSC= Mobile Telephone Service Center, BTS = Base Transceiver Station
HLR=Home Location Register, VLR=Visiting Location Register
2G CDMA Cellular
IS-95 is the best known example of 2G with CDMA
Advantages of CDMA for Cellular
• Frequency diversity – frequency-dependent
transmission impairments have less effect on signal
• Multipath resistance – chipping codes used for CDMA
exhibit low cross correlation and low autocorrelation
• Privacy – privacy is inherent since spread spectrum is
obtained by use of noise-like signals
• Graceful degradation – system only gradually degrades
as more users access the system
2G Technologies
cdmaOne (IS-95) GSM, DCS-1900 IS-54/IS-136
PDC
Uplink Frequencies 824-849 (Cellular) 890-915 MHz (Eurpe) 800 MHz, 1500 Mhz
(MHz) 1850-1910 (US PCS) 1850-1910 (US PCS) (Japan)
1850-1910 (US PCS)
Downlink Frequencies 869-894 MHz (US 935-960 (Europa) 869-894 MHz (Cellular)
Cellular) 1930-1990 (US PCS) 1930-1990 (US PCS)
1930-1990 MHz (US 800 MHz, 1500 MHz
PCS) (Japan)
Deplexing FDD FDD FDD
Multiple Access CDMA TDMA TDMA
Modulation BPSK with Quadrature GMSK with BT=0.3 /4 DQPSK
Spreading
Carrier Seperation 1.25 MHz 200 KHz 30 KHz (IS-136)
(25 KHz PDC)
Channel Data Rate 1.2288 Mchips/sec 270.833 Kbps 48.6 Kbps (IS-136)
42 Kbps (PDC)
Voice Channels per 64 8 3
carrier
Speech Coding CELP at 13Kbps RPE-LTP at 13 Kbps VSELP at 7.95 Kbps
EVRC at 8Kbps
Alternatives to 3G Cellular
• Major technical undertaking with many organizational and
marketing overtones.
• Questions about the need for the additional investment for
3G (happy with 2.5G)
• Wireless LAN in public places such as shopping malls and
airports offer options
• Other high-speed wireless-data solutions compete with 3G
• Mobitex low data rates (nominally 8 Kbps), it uses a narrowband (2.5KHz) as
compared to 30 KHz (GSM) and 5 MHz (3G).
• Ricochet: 40 -128 kbps data rates. Bankruptcy
• Flash-OFDM: 1.5 Mbps (upto 3 Mbps)
4G Systems
• Wireless networks with cellular data rates of 20 Mbits/second
and beyond.
• AT&T has began a two-phase upgrade of its wireless network on
the way to 4G Access.
• Nortel developing developing features for Internet protocol-
based 4G networks
• Alcatel, Ericsson, Nokia and Siemens found a new Wireless
World Research Forum (WWRF) for research on wireless
communications beyond 3G.
• Many new technologies and techniques (multiplexing, intelligent
antennas, digital signal processing)
• Industry response is mixed (some very critical)
What is 4G?
• Fourth Generation Technology
• Faster and more reliable
• 100 Mb/s as Ideal Data Rate
• Lower cost than previous generations
• Multi-standard wireless system
• Bluetooth, Wired, Wireless
• Ad Hoc Networking
• IPv6 Core
• OFDM used instead of CDMA
• Potentially IEEE standard 802.11n
• Most information is proprietary
Communications Architecture
• Broadcast layer:
• fix access points, (i.e.) cell tower connected by fiber, microwave, or satellite
(ISP)
• Ad-hoc/hot-spot layer:
• wireless LANs (i.e. internet at Starbuck’s)
Communications Architecture
• Personal Layer Gateway:
• devices that connect to upper layers; cell phone, fax, voice, data modem, MP3
players, PDAs

• Info-Sensor layer:
• environmental sensors

• Fiber-optic wire layer:

• high speed subterranean labyrinth of fiber optic cables and repeaters
Ad Hoc Networks

• Spontaneous self organization

of networks of devices

• Not necessarily connected to

internet

• 4G will create hybrid wireless

networks using Ad Hoc
networks

• Form of mesh networking–

Very reliable
Enhance Mobile Gaming
• Experience enhance wireless capabilities that deliver mobile
gaming interaction with less than five seconds

• Play online multi player games while traveling at high speeds or

sitting outside
Broadband access in Remote location
• 4G will provide a wireless alternative for broadband access

• I will provide first opportunity for broadband access in remote

locations without an infrastructure to support cable or DSL access.