Small scale fading and multipath

Mobile Radio Propagation

LOS

NLOS

LOS - line of sight NLOS - non line of sight

● Power received by the receiver is most important issue

i.e., signal level at receiver
● Signal level is affected
by channel condition and mobility of the receiver. Because receiver receives
Transmitted signal with multi-path after multi-reflection with different lengths of path.
● Propagation model is required to predict average received signal strength
at a given distance from the transmitter.
 Large scale and small scale propagation model
Large scale propagation model: A propagation model which characterizes
received signal strength over large distance between transmitter and receiver
is called large scale propagation model. (several hundreds or thousands meters
small scale propagation model: A propagation model which characterizes
received signal strength over short distance between transmitter and receiver
is called small scale propagation model. ( few wave lengths or during few seconds)

Large scale fading

Received power(dBm)

-30
Local average signal level
-40
-50 over long distance from transmitter
-60
-70 Small scale fading
14 15 16 17 18
T-R separation meters Instantaneous signal level due to move
by fraction of wavelength

Our interest is to study large scale propagation model

 Small scaling fading

Multipath fading
Fading caused by interference between two or more versions of same transmitted signal
which arrive at the receiver at slightly different times is called multiparth fading.

Effects of small scale fading

● Rapid changes in signal strength over a small travel distance or time interval
● random frequency modulation due to varying doppler shifts on different multiparth signals

● Time disperson (echoes) caused by multipath propagation delays.

Factors affecting small-scale fading

● Multipath propagatin: Random phase and amplitude change lead to small scale fading and distortions of signals

● Speed of mobile: Dopler shift

● Speed of surrounding objects: introduces time varying Doppler shift

● The transmission bandwidth of the signal: channel bandwidth quantifies the coherence bandwidth
 Doppler Shift Geomerty
l  d cos 
 t.v. cos 
2
  .l

2 From(1) and(2)
 .v.t. cos  (1)

Doppler shift v
f d  cos 
1    
fd    (2)
2  t 
 Problem:

Transmitter frequency1850MHz v=60mph

Estimate received freqnency

(a)Moving difrectly towards transmitter   0.162m v  26.82m / s

26.82
f  f c  f d  1850  cos 0  1850.00016MHz
0.162

(b) Moving direct away from transmitter

26.82
f  fc  f d  1850   1849.999834 MHz
0.162

© Moving parpendicular to the angle of arrival of the transmitted signal

  90
cos   0
No doppler shift
 Impulse response model
Small scale variations Vs impulse response of the channels
Mobile radio channel can be modeled as a linear filter with a time varying impulse response
h(d,t)

Transmitter o/p is the input x(t) Linear filer Received signal is the output
y(d,t)
Channel

Impulse response is a useful characterization of the channel, since it may be used to predict and compare
The performance of many different mobile communication systems and transmission bandwidths for
A particular mobile channel condition

d
d=vt

y (d , t )  x(t )  h(d , t )   x( )h(d , t   )d y (d , t )  t  x( )h(d , t   )d y (vt )  t  x( )h(vt , t   )d

y (t )  t  x( )h(vt , t   )d

Let us assume
x(t ) - Transmitted bandpass waveform

y (t ) - Received waveform
h(t , ) - Impulse response of time varying multipath radio channel
t is the time variatios due to motion
 is the channel multipath delay for a fixed value to t
h(t , )

x(t) Linear filer y(t) y (t )  x(t )  h(t , )

Channel
Bandpass channel response model
If c(t) and r(t) are complex envelopes of x(t) and y(t), i.e.,
x(t )  Re c(t ) exp( j 2f c t ) y (t )  Re r (t ) exp( j 2f c t ) and h(t , ) 
1
2

Re hb (t , )e jct 
1
hb (t ,  )
2
1
Linear filer r(t) r (t )  c(t )  hb (t , )
c(t) 2
Channel

Baseband equivalent channel

2
It is proved that x 2 (t )  0.5 c(t )
Response model
The factor ½ is due to the properties of the complex envelope
Excess delay axis is discretized into equal equual time delay segments which are called excess delay bins.


….
0 1 i  i 1


   i 1   i where 0  0

For i=0,    1   0 Frequency span=

2

 Since received signal in a multipath channel consists of a series of attenuated, time-delayed, phase
shifted replicas of the transmitted signal, the baseband impulse response of a multipath channel can be
expressed as,

N 1
hb (t , )   ai (t , ) exp[ j (2f c i (t )  i (t , ))] (   i (t ))
i 0
ai (t , ) and  i (t ) Are real amplitudes and excess delays, respectively, of ith multipath component at time t.

2f c i (t )  i (t , ) Phase shift due to free space propagation of the ith multipath complement, plus any additional
Phase shifts which are encountered in the channel.

t3  (t3 )
t2
 (t 2 )
t1
 (t1 )

t0
 (t 0 )
0 1 2 3 4
 If channel impulse response is time invariance

N 1
hb ( )   ai exp( j i ) (   i )
i 0

Probe pulse p (t )   (t   ) Is used to sound the channel to determine hb ( )

Example 5.2:
In urban RF radio channel,  max  100s (a) Find  ?
 max
In microcellular channel,  max  4s    1.5625us
N
Number of multipath bins, N  64
(b) Find maximum RF bandwidth which two models (microcell and microcell) can accurately represent in two

2
Maximum RF bandwidth that the model can accurately represent,   1.28MHz (macrocell)

 max 4
In urban microcell,     62.5ns
N 64
2 2
Maximum RF bandwidth   32MHz
 62.5ns
© In indoor propagation,  max  500ns What is RF bandwidth supported?

 max 500 109

    7.8125ns
N 64
RF bandwidth  2  2
 256 MHz
 7.8125ns
 Bandwidth Vs Received Power
The impulse response of multipath channel is measured in the field
Using channel sounding techniques.

Small-scale fading differ two signals with different bandwidths in

Identical multipath channel.

Consider a pulse RF (transmitted)

x(t )  Re p (t ) exp( j 2f ct )

Tbb is very narrow

Trep   max

 max Rect(t)
Assume, p (t )  2 for 0  t  Tbb
Tbb
 max
2
 max T Tbb
p(t )  2 rect (t  bb )
Tbb 2
Tbb t
Recall the low pass channel filter output
1
r (t )  c(t )  hb (t , )
2
N 1
hb ( )   ai exp( j i ) (   i ) For time invariant
i 0

c(t)=p(t)
1
r (t )  p (t )  hb ( )
2

1 N 1
r (t )  p(t )   ai exp( ji ) (   i )
2 i 0
1 N 1
  ai exp( ji ) p(t   i )
2 i 0
N 1  max  Tbb 
  ai exp( ji ) rect t  i 
i 0 Tbb  2 
Received power at any time t0 | r (t0 ) |2

1  max  1 N 1 
2 2 2
| r (t0 ) |     ak (t0 ) p (t   k )  dt
 max 0  4 k  0 
 2
1 1 N 1 2 max 
   T 

  ak (t0 )  2 max rect t  bb   i   dt
 max 4 k  0 0   Tbb  2 

N 1 2  max 1
 T 
  ak (t0 )  rect t  bb   i dt
k 0 0 Tbb  2 
N 1 Tbb 1
2
  ak (t0 )  dt
k 0 0 Tbb
N 1 2
  ak (t0 )
k 0
 Total received power is the sum the powers in individual multipath components

The received power from multipath components form random process

So, each component has a random amplitude and phase at any time t

The average small-scale received power for the wideband probe is given as follows

N 1
Ea, ( PWB )   ak 2 (t0 )
k 0

Average small-scale received power is the sum of the average powers received in
Each multipath component
 What happens if instead of pulse signal a CW signal is transmitted into the
Exact same channel
Assume c(t)=2
Instantaneous envelope of received signal is given by
N 1
r (t )   ai exp( ji (t , )
i 0
N 1
Instantaneous power is given by | r (t ) | |  ai exp( ji (t , )) |2
2
i 0

Received power is affected by the fluctuation of ai and i

ai Varies little over local areas
i Changes in propagation distance over space resulting large fluctuation of r(t)
r(t) is phasor sum of individual multipath compoent
r(t) fluctuates even when receiver moves over a small distance in the order of wavelength
Average received power

 N 1 j i 2 
Ea , [ PCW ]  Ea , |  ai e | 
 i 0 

N 1 j  j i 
    ai e i 

 ai e 
i 0   

N 12 N 1 N
  ai  2   rij cos(i   j )
i 0 i 0 i, j i