Wireless communications

Wireless channel propagation

characteristics

Dr. Ali Mohydeen

Spring 2024

1
Outline
• Introduction
• Fading in wireless channels
• Large scale fading
- Path loss and shadowing
• Small scale fading
- Flat fading and frequency selective fading
- Power delay profile and coherence bandwidth
- Doppler spread and coherence time

2
Introduction
• A major obstacle to building a reliable high speed wireless communication system
is the wireless propagation medium.

• Establishing a reliable wireless communication system requires deep

understanding of wireless channel propagation models and characteristics.

• Several factors are involved in the process of determining the channel behavior:
signal/channel bandwidth, environment or propagation medium, noise, etc.

• Due to these factors, a signal transmitted through the environment exhibits

fluctuation and attenuation in its level at the receiver side.

3
 Fading in wireless channels
• Two types of fading characterize a
wireless channel: large scale fading and
small scale fading.

• Large scale fading refers to the signal

power attenuation and fluctuation due to
path loss and shadowing. Path loss
refers to power loss due to the
propagation over large distances.

• Shadowing is when signals are

interrupted or blocked by large objects
such as mountains and buildings over
the propagation path between the
transmitter and the receiver.

4
Fading in wireless channels
• Path loss, shadowing and multipath fading Multipath propagation

5
Path loss
• The ratio of received signal power to transmitted signal power in free space is given as
follows (Friis model) :

𝑃𝑟 λ 2
= 𝐺𝑡 𝐺𝑟
𝑃𝑡 4𝜋𝑑

where 𝑃𝑡 and 𝑃𝑟 are the transmitted and received signal powers respectively, 𝐺𝑡 and 𝐺𝑟 are
the transmit and receive antenna gains respectively, λ is the wavelength, and 𝑑 is the
distance between the transmit and receive antennas.

• The free space path loss is expressed as:

2
𝑃𝑡 λ
𝑃𝐿dB = 10log10 = −10log10 𝐺𝑡 𝐺𝑟
𝑃𝑟 4𝜋𝑑

6
Path loss
• Example: A point-to-point link uses a transmit power of 1 Watt, transmit and
receive antennas with gains of 20 dB and operates at 3 GHz. How much power is
received by a receiver 300m away?

7
Path loss
• A more generalized form of path loss model:
𝑑
𝑃𝐿dB 𝑑 = 𝑃𝐿dB 𝑑0 + 10𝑛log10
𝑑0
where 𝑃𝐿dB 𝑑 is the average path loss at distance 𝑑, 𝑑0 is a reference distance
(close-in reference distance) and 𝑃𝐿dB 𝑑0 is the reference path loss.

• The parameters 𝑑0 , 𝑃𝐿dB 𝑑0 and 𝑛 depend on the antenna characteristics and the
propagation environment and can be obtained based on field measurements.
4𝜋𝑑0
• In free space, 𝑃𝐿dB 𝑑0 = 20log10 (for unity gain antennas) and 𝑛 = 2.
𝜆

8
Path loss and shadowing
• Including the effect of shadowing, the path loss at a distance 𝑑 is given as:
Shadowing
𝑃𝐿dB 𝑑 = 𝑃𝐿dB 𝑑 + 𝑆dB effect
𝑑
where 𝑃𝐿dB 𝑑 = 𝑃𝐿dB 𝑑0 + 10𝑛log10 𝑑0
is the average path loss and 𝑆dB is a Gaussian random
variable with a mean of 0 dB and variance 𝜎𝑆2dB (𝑆dB = 10log10 𝑆 , 𝑆 is a lognormal random variable),
the model is known as log-normal shadowing model.

• The received power at distance 𝑑 can be written as:

𝑃𝑟 (𝑑)dBm = 𝑃𝑡 dBm − 𝑃𝐿dB 𝑑 = 𝑃𝑟 (𝑑)dBm − 𝑆dB
where 𝑃𝑡 dBm is the transmitted power in dBm and
𝑑
𝑃𝑟 (𝑑)dBm = 𝑃𝑡 dBm − 𝑃𝐿dB 𝑑0 − 10𝑛log10
𝑑0
is the average received power at distance 𝑑.

9
Outage probability under path loss and shadowing

• Outage probability under path loss and shadowing is the probability that the received
power at a distance 𝑑 is less than the minimum acceptable received power level 𝑃min :

𝑃out 𝑑 = 𝑃 𝑃𝑟 (𝑑) < 𝑃min = 𝑃 𝑆dB > 𝑃𝑟 𝑑 − 𝑃min

𝑃𝑟 𝑑 −𝑃min 𝑃min −𝑃𝑟 (𝑑)

=𝑄 =1−𝑄
𝜎𝑆dB 𝜎𝑆dB

1 ∞ 𝑢2
• The Q-function is defined as: 𝑄 𝑥 = exp(− )𝑑𝑢.
2𝜋 𝑥 2

10
Outage probability under path loss and shadowing

• Example: In a wireless communication system, the receiver sensitivity (minimum

acceptable power level) is -110 dBm, the average received signal power at a
distance 𝑑 is -100 dBm. The variance of lognormal shadowing 𝜎𝑆2dB is 25 dB. Find
the probability of outage at distance 𝑑.

11
Link Budget
• The link budget is an important tool for designing wireless communication systems.

• A link budget provides an evaluation of system performance by investigating the different gains
and losses that may be experienced by signals in a wireless communication system.

• It allows predicting the received signal strength along with the appropriate link margin.

• The link margin is the excess amount of power above the minimum acceptable received power
level (receiver sensitivity), allocated to account for different possible losses, ensuring more reliable
communication.

• The link margin allocated to compensate for fading is sometimes referred to as the fade margin.

12
Small scale fading
• Small scale fading refers to the rapid fluctuation of the signal over short periods or
short travel distances.

• The small scale fading scheme can be divided into two categories. The first one is
related to the multipath delay spread, and the second is related to the Doppler
spread.

13
Small scale fading

• Transmitted radio signal:

𝑠 𝑡 = 𝑅𝑒{𝑠𝑏 (𝑡)𝑒 𝑗2𝜋𝑓𝑐𝑡 }
𝑠𝑏 (𝑡) is a complex baseband signal.

• Received signal:
𝑟 𝑡 = 𝑅𝑒{{ 𝐿𝑙=1 𝑎𝑙 𝑠𝑏 𝑡 − 𝜏𝑙 𝑒 −𝑗2𝜋𝑓𝑐𝜏𝑙 } 𝑒 𝑗2𝜋𝑓𝑐𝑡 }
𝑎𝑙 is the attenuation of path 𝑙 and 𝜏𝑙 is the corresponding time delay.

• The complex baseband received signal:

𝑟𝑏 𝑡 = 𝐿𝑙=1 𝑎𝑙 𝑠𝑏 𝑡 − 𝜏𝑙 𝑒 −𝑗2𝜋𝑓𝑐𝜏𝑙

14
Flat fading

Narrowband assumption: 𝑠𝑏 𝑡 − 𝜏𝑙 ≈ 𝑠𝑏 (𝑡) Power fluctuation

(Symbol duration is much greater than channel delay spread)

𝐿 −𝑗2𝜋𝑓𝑐 𝜏𝑙
Hence 𝑟𝑏 𝑡 = 𝑠𝑏 𝑡 𝑙=1 𝑎𝑙 𝑒
ℎ

The channel is seen as a single coefficient

h. CIR is a single impulse. Depending on the values of the time delays 𝜏𝑙 , the
different propagation paths can add up
constructively or destructively causing fading.
15
Flat fading
Statistics of the fading coefficient

• The fading coefficient is given as:

𝐿 −𝑗2𝜋𝑓𝑐 𝜏𝑙
ℎ= 𝑙=1 𝑎𝑙 𝑒
which can be written as:
ℎ = 𝑋𝑟 + 𝑗𝑋𝑖

𝐿 𝐿
where 𝑋𝑟 = 𝑙=1 𝑎𝑙 cos 2𝜋𝑓𝑐 𝜏𝑙 and 𝑋𝑖 = − 𝑙=1 𝑎𝑙 s𝑖𝑛 2𝜋𝑓𝑐 𝜏𝑙 .

• Each of 𝑋𝑟 and 𝑋𝑖 is the sum of a large number of random components.

• According to central limit theorem, 𝑋𝑟 and 𝑋𝑖 can be modeled as Gaussian random variables.

16
Flat fading
Statistics of the fading coefficient
ℎ = 𝑋𝑟 + 𝑗𝑋𝑖 = 𝑎𝑒 𝑗𝜙
𝑋𝑖
where 𝑎 = 𝑋𝑟2 + 𝑋𝑖2 is the magnitude and 𝜙 = tan−1 .
𝑋𝑟

• Assuming that 𝑋𝑟 and 𝑋𝑖 are zero mean independent random variables with same variance
𝜎𝑋2 , the magnitude coefficient is Rayleigh distributed with following probability density:
−𝑎2
𝑎 2𝜎2
𝑓𝑎 𝑎 = 2 𝑒 𝑋
𝜎𝑋
where 𝜙 is uniformly distributed over −𝜋, 𝜋 .

17
Flat fading
Statistics of the fading coefficient
Rayleigh probability density function

• The Rayleigh fading is

reasonable when there is no
line of sight path between the
transmitter and receiver.

• When the line of sight path

exists, Rician fading becomes
a more reasonable model.

18
Bit error rate (BER)
• The probability of bit error bit error rate (BER) of a BPSK system over an additive white
Gaussian noise (AWGN) channel is given as:
2𝐸𝑠 1 𝐸𝑠 1 𝐸𝑏
𝐵𝐸𝑅 = 𝑄 = erfc = erfc
𝑁0 2 𝑁0 2 𝑁0

Where 𝐸𝑠 is the symbol energy, or bit energy (𝐸𝑏 ) (in BPSk, a symbol is one bit), 𝑁0 is the
1 ∞ 𝑢2
Noise power spectral density and 𝑄 𝑥 = exp(− )𝑑𝑢.
2𝜋 𝑥 2
𝐸𝑏
• is the ratio of energy per bit to the noise power spectral density and it can be referred to
𝑁0
as the signal to noise ratio (SNR) per bit. It is the signal to noise ratio for a digital
communication system.

19
Bit error rate (BER)
Bit error rate (BER) for a Rayleigh fading channel (with AWGN noise)

• For a Rayleigh fading channel, the BER is given as follows:

1 ℎ 2 𝐸𝑏 ℎ is random
BER fading = 𝑒𝑟𝑓𝑐
2 𝑁0
• The average BER for a Rayleigh fading channel is then given as:
∞ 𝐸𝑏
1 ℎ 2 𝐸𝑏 1 𝑁0
BER fading = 𝑒𝑟𝑓𝑐 𝑓𝑎 𝑎 𝑑𝑎 = 1−
2 𝑁0 2 𝐸
0 1+ 𝑏
𝑁0

20
Bit error rate (BER)
• Example: For a given BPSK communication system, the SNR per bit 𝐸𝑏 𝑁0
= 10 𝑑𝐵.
1) What is the Bit error rate over an AWGN channel?
2) What is the Bit error rate over a wireless Rayleigh fading channel?

21
Frequency selective fading
• A wideband signal whose bandwidth is greater than the channel coherence
bandwidth will experience frequency selective fading.
• Some path delays can now be resolvable.

Channel impulse
response

22
Frequency selective fading
• The time varying channel impulse response (CIR) is given as:
𝐿

ℎ 𝑡, 𝜏 = 𝛼 𝑙 𝑡 𝛿 𝜏 − 𝜏𝑙 𝑡
𝑙=1

• For time invariant channel:

𝐿

ℎ 𝜏 = 𝛼 𝑙 𝛿 𝜏 − 𝜏𝑙
𝑙=1

where L is the number of resolvable paths and 𝛼𝑙 is the corresponding gain parameter
including phase parameter.

23
Power delay profile
• The power delay profile (PDP) is an important parameter for characterizing frequency-selective
fading channels.

• PDP is given as the measure of the received signal power as a function of the channel propagation
delays, it is defined as:
𝑃𝑑 𝜏 = 𝐸 𝛼𝑙 2 𝛿(𝜏 − 𝜏𝑙 )
where 𝐸 𝛼𝑙 2 is the power associated to path delay 𝜏𝑙 .

• The main parameter characterizing the power delay profile are the mean delay, the root mean
square (RMS) delay spread, and the maximum excess delay (or maximum delay spread).

24
Power delay profile
• The mean delay is given as:

𝐿
𝑙=1 𝜏𝑙 𝑃𝑑 (𝜏𝑙 )
𝜏= 𝐿
𝑙=1 𝑃𝑑 (𝜏𝑙 )
• The RMS delay spread is given as:
𝜎𝜏 = 𝜏2 − 𝜏2
𝐿 2
𝑙=1 𝜏𝑙 𝑃𝑑 (𝜏𝑙 )
where 𝜏 2 = 𝐿 𝑃 (𝜏 ) .
𝑙=1 𝑑 𝑙

• The maximum excess delay is given as:

𝜏𝑚𝑎𝑥 = 𝜏𝐿 − 𝜏1

25
Coherence bandwidth
• Coherence bandwidth (𝐵𝑐 ) is a key parameter to characterize a wireless channel.

• The coherence bandwidth is the frequency range over which the channel frequency response is almost flat.

• Coherence bandwidth is inversely proportional to the channel delay spread 𝜎𝜏 .

• The expression of the channel coherence bandwidth in terms of the delay spread is based on the channel
frequency domain auto-correlation function.
• If the coherence bandwidth is defined as the frequency range over which the frequency domain auto-
correlation function is at least 0.9, then it is given as:
1
𝐵𝑐 =
50𝜎𝜏
• If the coherence bandwidth is defined as the frequency range over which the frequency domain auto-
correlation function is at least 0.5, then it is given as:
1
𝐵𝑐 =
5𝜎𝜏

26
Coherence bandwidth
• The value of the channel coherence bandwidth compared to the signal bandwidth, provide
information about the type of fading experienced.

• When the bandwidth of the transmitted signal (𝐵𝑠 ) is less than the channel coherence bandwidth,
the resulting phenomenon is flat fading, while when 𝐵𝑠 > 𝐵𝑐 , the fading is frequency selective.

27
Power delay profile – coherence bandwidth
• Example 1: For the following power delay profile:

0 dB

- 10 dB

- 20 dB

0 1 3 5 𝜇𝑠

1) Find the mean delay and RMS delay spread.

2) Find channel coherence bandwidth.

28
Power delay profile – coherence bandwidth
1
• Example 2: Consider that the coherence bandwidth is given by 𝐵𝑐 ≈ . Show that a flat fading
5𝜎𝜏
channel occurs when 𝑇𝑠 > 10𝜎𝜏 . Note that 𝐵𝑐 is the RF bandwidth, and assume that 𝑇𝑠 is the
reciprocal of the baseband signal.

29
InterSymbol Interference (ISI)
• The main impairment introduced in the case of frequency selective
fading is intersymbol interference (ISI).

A B C D

A B C D

30
Doppler spread
• Doppler effect (or Doppler shift/spread) refers to the change in the frequency of the propagating
electromagnetic wave due to the relative motion between the transmitter and the receiver, or due to
the mobility of scattering objects in the environment.

• Consider a mobile user (considered as transmitter) traveling through a base station (receiver) at a
speed 𝜗. Due to the Doppler shift, the frequency received at the base station is given as:

𝑓𝑟 = 𝑓𝑐 + 𝑓𝑑
where 𝑓𝑐 is the carrier frequency and 𝑓𝑑 the Doppler frequency shift given as:

𝜗 cos 𝜃
𝑓𝑑 = 𝑓𝑐
𝑐
where 𝑐 is the speed of light and 𝜃 is the angle between the direction of motion and the base station.

31
Doppler spread
• The time varying narrowband (flat fading) coefficient is then given as:
𝐿

ℎ 𝑡 = 𝑎𝑙 (𝑡)𝑒 −𝑗𝜙𝑙 (𝑡)

𝑙=1
where 𝜙𝑙 (𝑡) is the phase parameter that includes the contribution of Doppler in addition to
the phase due to the propagation delay.

• Assuming that the amplitude, delay and Doppler frequency of a given channel path 𝑙 are varying slowly
enough over time intervals of interest, the time varying channel coefficient can be written as:
𝐿

ℎ 𝑡 = 𝑎𝑙 𝑒 −𝑗𝜙𝑙 (𝑡)
𝑙=1
where
𝜙𝑙 𝑡 = 2𝜋𝑓𝑐 𝜏𝑙 − 2𝜋𝑓𝑑𝑙 𝑡

32
Doppler spread and coherence time
• The doppler spread is sometimes called the channel fading rate or fading bandwidth.

• The key parameter related to Doppler effect is the channel coherence time (𝑇𝑐 ).

• The channel coherence time is the time interval over which the channel is almost invariant
(static).

• 𝑇𝑐 is inversely proportional to 𝑓𝑑𝑚𝑎𝑥 , where 𝑓𝑑𝑚𝑎𝑥 is the maximum Doppler shift given
as:
𝜗
𝑓𝑑𝑚𝑎𝑥 = 𝑓𝑐
𝑐

33
Doppler spread and coherence time
• The channel coherence time can be evaluated based on the channel time domain
autocorrelation function defined as:

𝑅ℎℎ 𝑡, 𝑡 + 𝛥𝑡 = 𝐸 ℎ(𝑡)ℎ∗ (𝑡 + 𝛥𝑡)

• Assuming uniform scattering (Jakes’ model), the channel autocorrelation function can be
expressed as follows:

𝑅ℎℎ 𝑡, 𝑡 + 𝛥𝑡 = 𝐸 ℎ 2 𝐽0 2𝜋𝑓𝑑𝑚𝑎𝑥 𝛥𝑡 = 𝑅ℎℎ 𝛥𝑡

where 𝐽0 is the zero-order Bessel function.
• The autocorrelation function depends only on 𝛥𝑡, the channel response is wide sense
stationary (WSS).

34
Doppler spread and coherence time
• In the frequency domain, the Doppler power spectrum of the channel is given as:

+∞
𝑆 𝑓 = 𝑅
−∞ ℎℎ
𝛥𝑡 𝑒 −𝑗2𝜋𝑓𝛥𝑡 𝑑𝛥𝑡
1
2
, if |𝑓| ≤ 𝑓𝑑𝑚𝑎𝑥
𝑓
= 𝜋𝑓𝑑𝑚𝑎𝑥 1−
𝑓𝑑
𝑚𝑎𝑥
0 otherwise

• Due to the Doppler shift, the propagating signals exhibit what is called “spectral
broadening”.
35
Doppler spread and coherence time
• If coherence time 𝑇𝑐 is defined as the time interval over which the channel autocorrelation
function 𝑅ℎℎ 𝛥𝑡 is greater than 0.5, then 𝑇𝑐 is given as:

9
𝑇𝑐 ≈
16𝜋𝑓𝑑𝑚𝑎𝑥

• Based on the value of 𝑇𝑐 relative to the symbol period 𝑇𝑠 , we have two types of fading:
slow fading and fast fading.
• The fading phenomenon is referred to as slow fading when the transmitted symbol period
is less than the channel coherence time, that is 𝑇𝑠 < 𝑇𝑐 (𝐵𝑠 > 𝐵𝐷 ), while when 𝑇𝑠 > 𝑇𝑐 (𝐵𝑠
<𝐵𝐷 ), it is referred to as fast fading.

36
Doppler spread and coherence time
• Example: Suppose a wireless transmission is carried out in a fading channel. The time duration of
the symbol is 𝑇𝑠 = 100𝜇𝑠, carrier frequency is 800 MHz, the receiver is moving at a speed of
40km/h.

1) Calculate the 50% coherence time of the channel.

2) Specify whether the system is in slow fading or fast fading.

37
Classification of small scale fading

38
Level crossing rate and average fade duration
• Level crossing rate (LCR) and average fade duration (AFD) are two important parameters that allow
understanding the time varying behavior of a wireless channel.

• Level crossing rate is the rate at which the received signal crosses a given threshold. Under Rayleigh fading,
the level crossing rate is given as:
2
LCR = 2𝜋𝑓𝑑𝑚𝑎𝑥 𝜌𝑒 −𝜌
where 𝜌 is the threshold normalized to the root mean square (RMS) signal level.

• Average fade duration is a measure of the average time the signal remains below a given threshold. Under
Rayleigh fading, the average fade duration is given as:
2
𝑒 𝜌 −1
AFD = 𝜌𝑓
𝑑𝑚𝑎𝑥 2𝜋

39