Chapter 04- Small Scale Fading

Small-scale Fading – Multipath

• Small-scale fading (or simply fading) describes rapid
fluctuation of amplitude of a radio signal over
• Short period of time or
• Small travel distance
• Is severe than the large-scale path loss
• Fading is caused by multipath (self) interference
• Multipath: Two or more versions of a transmitted signal
• Multipath signals, if arrive at slightly different times, may combine
at the receiver antenna distractively that causes signal fluctuation

2
Small-scale Fading – Mobility
• Effects of multipath
• Time dispersion (echoes) caused by multipath propagation delays
• Frequency selectivity nature of the wireless channel as a result of
time dispersion

• Another source of fading is mobility

• Random frequency dispersion (modulation) resulting from Doppler
shifts on different multipath signals (Doppler shift is b/c of mobility)
• As a result of mobility, the channel becomes time varying

• Fading depends on
• Relative propagation time of the signals
• Bandwidth of the transmitted signal (and data rate)
• Speed of receiver and surrounding object

3
Small-scale Fading – Influencing Factors
1. Multipath propagation
• Presence of reflecting objects and scatterers causes
• Multiple version of the signal to arrive at the receiver with different
amplitude and time delays
• Relevant terms: Delay spread and coherence bandwidth

2. Speed of mobile
• Causes Doppler shift (“+” or “-”) at each multipath component
• Results in random frequency modulation
3. Speed of surrounding objects
• A receiver moving at high speed can pass through several fades in
small period of time

• Causes time-varying Doppler shift on the multipath components

• If the surrounding objects move at a greater rate than the mobile,
then this effect dominates the small-scale fading and vice versa

4
Small-scale Fading – Influencing Factors
• The term coherence time determines how “static” the channel is
(and depends on the Doppler shift), e.g., room environment ,
outdoor, urban, …

4. Bandwidth of the signal

• The channel bandwidth can be quantified by the term coherence
bandwidth, Bc
• Coherence bandwidth measures the maximum frequency difference
for which signals are still strongly correlated in amplitude

• If BW of the signal is greater than the coherence bandwidth, the

received signal will be distorted (filtered)
• However, the signal strength will not fade much over a local area (i.e.,
small-scale fading will not be significant)
• If the transmitted signal has a narrow bandwidth as compared to
the channel, signal will not be distorted in frequency
• Concept of frequency diversity?
5
Doppler Shift – Illustration
• Phase change in the received
signal due to path difference
2πΔl 2πvΔt
Δ = = cosθ
λ λ
• The apparent change in frequency
1  v
fD =  = cosθ
2 t λ
• Relates the Doppler shift to the
mobile velocity
• Shift can be “+” or “-” depending
direction of motion

• Doppler spreading =>increasing

the signal bandwidth

6
Doppler Shift – Illustration
• When a wave source (transmitter) and/or a receiver is/are
moving, the frequency of the received signal will not be the
same as that of the transmitted signal
• When they are moving towards each other, the frequency of the
received signal is higher than the source
• When they are opposing each other, the frequency decreases

• Thus, the frequency of the received signal is

f R = fC  f D

• Where f C is the frequency of the source carrier

f D is the Doppler shift in frequency

7
Impulse Response Model
• Small-scale variations of a signal is related to the impulse
response of the mobile radio channel
• The impulse response is
• A wideband channel characterization
• Contains all information necessary to simulate or analyze any type
of channel

• A wireless channel can be modeled as a linear time

varying (LTV) filter
• The time variation is due to receiver motion in space
• We use discrete-time impulse response model

• Filtering is caused by the summation of amplitudes and

delays of the multipath signals at any instant of time

9
Impulse Response Model …
• Consider the transmission of a band-pass signal sb(t)

• where s(t) = x(t) + jy(t) is the complex envelop or complex

baseband signal
• X(t): In-phase component x(t)
• Y(t): Quadrature component

• Advantage of baseband representation

• It allows signal manipulation via s(t) irrespective of the carrier
frequency fc

10
Impulse Response Model …
• The received signal rb(t) will have a similar form

• Both rb(t) and r(t) depend on the channel’s behavior

• In multipath channel, the received signal is the sum of

• Line-of-sight path
• All resolvable multipath components

• Associated with each path is

• A propagation delay k(t)
• An attenuation factor k(t)
• Dk : Doppler phase shift associated with kth multipath component
• Propagation delay, attenuation factor, and Doppler shift are
time-variant as a result of time-varying nature of the channel
11
Impulse Response Model …
• Thus, the received bandpass signal may be expressed as

• Then substituting for sb(t), the received signal will be

• where
• N(t): the number of resolvable multipath components
• k(t): attenuation of the kth multipath component
• k(t): delay of the kth multipath component
• Dk : Doppler phase shift associated with kth multipath component

12
Impulse Response Model …
• The received low-pass equivalent signals is simplified as

Number of Overall phase shift

resolvable paths Attenuation (Doppler and delay)
Delay

• Where the overall phase shift is given as

• Lowpass equivalent impulse response of the channel,
written as a linear time-variant filter, takes the form

13
Impulse Response Model …
• Resolvable components
• Two multipath components, having delays 1 and 2, are resolvable if
the delay difference exceeds symbol duration Ts, i.e., 1 - 2 >> Ts

• Non-resolvable components
• Combined into a single multipath component with delay   1  2
• Cannot be separated at the receiver, since s(t - 1)  s(t - 2)

• The kth non-resolvable multipath component corresponds to

• The path associated with a single reflector or
• With multiple reflectors clustered together that generate multipath
components with similar delays

• Given lk(t) as the kth path length

• The delay of this path: k(t)= lk(t)/c
• The phase change associated with delay k(t):

14
Impulse Response Model - LTV Filters
• In the LTV model h(,t)
• The t represents the time variations due to motion
• The  represents the channel multipath delay for a fixed value of t

• Impulse response of a LTV filter h(,t) is the channel output

at t when the channel input is an impulse applied at t - 

• h(,t) is a function of two time variables

1. The instant when the impulse is applied to its input (initial time)
2. The instant of observing the output (final time)

15
Impulse Response Model - LTV Filters

16
Impulse Response Model - LTV Filters
• Example of time varying discrete-time impulse response
model for a multipath radio channel
• 0 represents the first arriving signal at the receiver & equal to zero
•  = 1 - 0 is the time delay

17
Bandwidth and Received Power Relation
• Power delay
profile
• In actual wireless
communication
systems, the
impulse response
of a multipath
channel is
measured in the
field using
channel sounding
techniques

18
Power Delay Profile
• Many multipath channel parameters are derived from the
power delay profile, (PDP)
• PDP is found by averaging instantaneous power delay profile
measurements over a local area

• PDPs are represented as plots of relative received power

as a function of excess delay with respect to a fixed time
delay reference

19
Measure Multipath Power Delay Profile (PDP)

The signal from

intermediate
reflectors
The signal from
far away
reflectors

From a 900 MHz cellular system in

San Francisco
The signal from
close by reflectors Inside a grocery store at 4 GHz

20
Power Delay Profile – Discrete Model

From a 900 MHz cellular system in

San Francisco

21
Delay Spread
• Each multipath signal travels different path length, so that
the time of arrival for each path is different
• A single transmitted pulse will be time spread (dispersion)
when it reaches the receiver
• This effect, which spreads out the signal, is called delay
spread
• Delay spread is a property of the communication channel
that may cause inter-symbol interference (ISI)
• Require channel equalizers, pulse shaping filters, rake receivers

22
Time Dispersion Parameters
• Mean excess delay: First moment of the PDP & defined as
  2
k k  P( ) k k
 = k
= k

 k
2
k  P( )k
k

• RMS delay spread: The square root of the second central

moment of the PDP
  =  − ( ) 2 2

• where   2
k
2
k  P( ) k
2
k
2 = k
= k

 k
2
k  P( ) k
k

• Maximum excess delay: The time delay during which

multipath energy falls to X dB below the maximum
• These delays are measured relative to the first detectable
signal arriving at the receiver at 0=0
23
Example: Indoor Power Delay Profile

Also called the

excess delay
spread

24
Typical Measured Values of RMS Delay Spread
Outdoor: Order of  sec.

Indoor: Order of n sec.

25
Example:

26
Coherence Bandwidth, BC
• Characterizes the channel in the frequency domain
• Analog to the delay spread parameter in the time domain

• Coherence bandwidth measures range of frequencies over

which the channel is considered to be “flat”
• Flat: A channel which passes all spectral components with
approximately equal gain and linear phases
• Is a statistical measure

• Or, coherence bandwidth defines frequency ranges in

which two freq components have a strong amplitude
correlation
• Two sinusoids with frequency separation greater than BC are
affected differently by the channel

27
Coherence Bandwidth …
• The rms delay spread and coherence bandwidth, BC, are
inversely proportional to each other
1. For frequency correlation function  0.9, then
1
BC 
50 
2. If the frequency correlation is relaxed to 0.5, then
1
BC 
5 
• Note: An exact relationship between BC and  does not
exist

28
Coherence Bandwidth – Example
• Consider the multipath power delay profile given in the
following figure
1. Calculate the mean excess delay and rms delay spread for the
2. Estimate the 50% coherence bandwidth of the channel
3. Would this channel be suitable for AMPS or GSM service without
the use of an equalizer?

29
Coherence Bandwidth – Solution

30
Doppler Spread
• Doppler spread and coherence time describe the time
varying nature of the wireless channel due to mobility
• Delay spread and coherence bandwidth describe the frequency
dispersive nature of the channel in a local area

• Mobility: Relative motion is either between MS and BS or

movement of objects in the channel
• Doppler spread, BD
• Measures the spectral broadening caused by the time rate of
change of the mobile radio channel

• Defined as the range of frequencies over which the received

Doppler spectrum is essentially non-zero

31
Doppler Spread …
• Let a sinusoidal tone of frequency fc is transmitted
• The received signal range spectrum, called the Doppler
spectrum, will have components in the range fc-fd to fc+fd,
where fd is the Doppler shift
• Doppler shift is a function of
• Relative velocity of the MS
• Frequency of the signal
• Angle  between the direction of motion of the MS and
direction of arrival of the scattered wave

• If the baseband signal bandwidth is much greater than BD,

the effects of Doppler Spread are negligible

32
Coherence Time
• Coherence Time, Tc: The time domain dual of BD
1
TC 
fm
• where: fm is the maximum Doppler Shift given by fm = v


• Coherence time
• Measures the time duration over which the channel impulse
response is invariant
• A statistical measure that quantifies the similarity (correlation) of
the channel at different times

• Two signals arriving with a time separation greater than Tc

are affected differently by the channel

33
Coherence Time …
• For time correlation function above 0.5, then
9
TC 
16f m
• A popular rule of thumb is to define the coherence time as
the geometric mean of the above two equations, i.e.,
9 0.423
TC = =
16f m2 fm
• E.g., for a vehicle traveling at 60 mph (26.8mps) using 900 MHz
carrier, a conservative value of TC is shown to be 2.22 ms
• If a digital system that transmits at symbol rate > 1/TC = 454bps, the
channel will not cause distortion due to motion

• However, distortion could result from multipath time delay spread,

depending on the channel impulse response

34
Coherence Time …
• Example: Determine the proper spatial sampling interval
required to make small-scale propagation measurements
which assume that consecutive samples are highly
correlated in time.
• How many samples will be required over 10 m travel
distance if fc = 1900 MHz and v=50m/s.
• How long would it take to make these measurements,
assuming they could be made in real time from a moving
vehicle?
• What is the Doppler spread BD for the channel?

35
Coherence Time – Solution

36
Types of Small-scale Fading
Small-scale Fading
(Based on Multipath Time Delay Spread)

Flat Fading Frequency Selective Fading

1. BW Signal < BW of Channel 1. BW Signal > BW of Channel
2. Delay Spread < Symbol Period 2. Delay Spread > Symbol Period

Small-scale Fading
(Based on Doppler Spread)

Fast Fading Slow Fading

1. High Doppler Spread

1. Low Doppler Spread
2. Coherence Time < Symbol Period 2. Coherence Time > Symbol Period
3. Channel variations faster than baseband 3. Channel variations smaller than baseband
signal variations
signal variations

37
Flat Fading
• The channel has a constant gain and linear phase
response over a BW > the BW of the transmitted signal
• The spectral characteristics of the transmitted signal are preserved
at the receiver
• Most common type of fading described in the literature
• However, the strength of the received signal change with
time, due to fluctuations in the gain of the channel caused
by the multipath
• Typically, flat fading channels cause deep fades, and may require
20 or 30 dB more transmitter power to achieve low bit error rates
• To summarize, a signal undergoes flat if

and

38
Flat Fading – Channel Characteristics
• The most common amplitude distribution is the Rayleigh
distribution

• The spectrum of the signal is preserved even if the

amplitude changes over time
39
Frequency Selective Fading
• Due to time dispersion of the transmitted symbols within
the channel
• It may introduce an intersymbol interference (ISI)

• Viewed in the frequency domain, certain frequency

components in the received signal spectrum have greater
gains than others
• A signal undergoes frequency selective fading if

and

40
Frequency Selective Fading – Channel Characteristics

41
Fading Due to Doppler- Fast Fading & Slow Fading
• Fast fading
• The channel impulse response changes rapidly within the symbol
duration, i.e.,
Bs< BD and Ts>Tc
• This causes frequency dispersion (also called time selective
fading) due to Doppler Spreading, which leads to signal distortion

• Slow fading
• The channel impulse response changes at a rate much slower
than the transmitted baseband signal
Ts<<Tc and Bs>>BD

42
Fading Channels
• Matrix illustrating type
of fading experienced
by a signal as a
function of
a) Symbol period
b) Baseband signal
bandwidth

43
Rayleigh Distribution
• Rayleigh distribution: Describes statistical time varying
nature of
• The received envelope of a flat fading signal or
• The envelop of an individual multipath component
• The envelop of the
sum of two
quadrature Gaussian
noise signals is
Rayleigh distribution
• A typical Rayleigh
fading envelop at
900 MHz is shown in
the figure

44
Rayleigh Distribution …
• Rayleigh distribution
has a pdf given by
 r  r2 
 2 exp − 2  (0  r   )
p ( r ) =   2 
0 (r  0)

• Note that
•  is the rms value of
the received signal
• 2 is the time-
average power of
the received signal

45
Rayleigh Distribution – Outage Probability
• Outage probability: The probability that the envelop of the
received signal does not exceed a specified value R is give
by the corresponding cumulative distribution function
(CDF) R
 R2 
P( R ) = Pr(r  R ) =  p (r )dr = 1 − exp − 
2 
0  2 

46
Ricean Distribution
• When there is a
dominant stationary
(non-fading) signal
component present,
such as a LOS
propagation path,
the small-scale
fading envelope
distribution is Ricean

47
Summary
• Small-scale fading composed of multipath & Doppler spread
• Multipath delay spread leads to time dispersion and frequency
selective fading
• Doppler spread leads to frequency dispersion and time selective
fading (the channel becomes time varying)
• The two fading mechanism are independent of one another
• Small-scale is modeled as a LTV filter and, accordingly,
channels are categorized as
• Flat, frequency selective, fast, and slow fading
• Fading parameters
• Delay spread, coherence bandwidth, Doppler spectrum, and
coherence time
• Various statistical fading models exist for hardware and
software simulation
48