Signaling over Wireless channel

Prepared by: Dr / Doaa Gamal

Assistant professor at Faculty of Engineering, Suez Canal University
([email protected])
 Basic Communication System
2

 Simple point-to-point communication system

𝑦 = 𝐻. 𝑥 + 𝑛
Noise model to mimic random nature processes: AWG
–Additive (superposition)
–Flat spectrum, i.e. white (uncorrelated random voltage signals)
–Gaussian distributed (multiple random voltage signals)
•Other noise models exist (Laplacian, uniform,…)
 AWGN
3
 Wireless channel
4
 Wireless channel
5

Wireless channel introduces:

 Noise (e.g. AWGN)

 Attenuation

 Distortion

 Interference
 Wireless channel
6

 Wave propagation depends on

the environment: buildings,
roads, trees and terrain.
 Each path will experience an
attenuation, a time delay and a
possible frequency shift
 The superposition of all the
signal paths produces a
distorted received signal.
 Wireless channel
7

Path Loss − caused by dissipation

of the power radiated by the
transmitter
−depends on the distance
between transmitter and receiver
Shadowing − caused by obstacles
between the transmitter and
receiver that absorb power
 Multipath
8

- Due to the presence of multiple random reflectors and scatterers during signal
propagation.
- Observed as multiple copies of the signal arriving at the receiver at different times and
with different power.
- The multiple copies of the transmitted signal, each having a different amplitude, phase
and delay are added at the receiver creating either constructive or destructive
interference.
- Due to the presence of motion at the transmitter, the receiver or the surrounding objects,
the multipath channel changes over time.
 Wireless fading channel
9
 Wireless fading channel
10
 Wireless fading channel
11
 Large scale fading
12

 Large-scale fading is caused and characterized by path

loss and shadowing.
 More important for cell site planning (calculate cell radius), less
for communication system design.
 Since variations in received power due to path loss and
shadowing occur over relatively large distances, these
variations are referred to as large-scale propagation
effects.
 Path loss models
13

 Deterministic Path-loss models : use analysis to

determine the relation between the received signal
power and the propagation distance .
 Empirical models: more sophisticated models (such as
okumura, and Hata models) are used, where the
received signal power attenuates with other more
parameters e.g., height and frequency)
 Deterministic Path loss models
14

Path Loss:

 Thus, the received signal power falls off in

 Inverse proportion to the square of the
distance 𝑑 between the transmit and
receive antennas.
𝑛
𝑃𝑟 1
𝛼
𝑃𝑡 𝑑
SMALL SCALE FADING
 Fading: Small Scale vs Large Scale
16
 Small scale fading
17

 Small scale fading (also called short-term fading and

multipath fading): rapid variations of the received
signal level in the short term
 Due to the presence of multiple random reflectors
and scatterers during signal propagation.
 Observed as multiple copies of the signal arriving at
the receiver at different times and with different
power.
 The multiple copies of the transmitted signal, each
having a different amplitude, phase and delay are
added at the receiver creating either constructive or
destructive interference.
 Small scale fading
18
 Small scale fading
19

𝑦𝑘 = 𝐴𝑖 sin 𝜔𝑡 + 𝜑𝑖 = 𝐴 sin 𝜔𝑡 + 𝜑
𝑖=1
MULTIPATH: : TIME-DISPERSION => FREQUENCY SELECTIVITY
 Multipath
21
 Multipath: : resolvable paths
22

• The received signals arriving from different paths

with very close delays can be approximated to a
single resolvable path with amplitude ℎ , phase 𝜃
and with delay 𝜏. This is re represented by delta-
derac function at 𝜏.
 Resolvable path
23
 Channel model
24

Discrete-time equivalent baseband

model of the channel:
Frequency selective (multipath)

ℎ𝑙 [𝑚] is the lth resolvable path at sample m

Single path(Flat fading)

 Multipath
25

 Channel maximum delay spread: the difference in

propagation time between the longest and shortest path,
counting only the paths with significant energy.
 A wideband signal with symbol duration Tc can "resolve"
the time dispersion of the channel with a maximum delay
spread of T delay with an accuracy of about Tc. the number
of resolvable paths is
 Multipath
26

 Power delay profile: function determined by the average

power associated with each path.
 Channel spectral response: Fourier transform of the power
delay profile.

Base-band equivalent
channel
 Multipath
27

The coherence bandwidth is the range of frequencies over which the

channel shows a flat frequency response.
 Multipath: : Time-Dispersion => Frequency Selectivity
28

 Narrow-band channels:
also called flat fading
(single tap channel)
 Wideband channels: also
called frequency
selective (multi-tap
channel)
 Multipath: : Time-Dispersion => Frequency Selectivity
29
 Multipath: : Time-Dispersion => Frequency Selectivity
30

 If the delay spread is longer than the symbol-duration, symbols will

“smear” onto adjacent symbols and cause symbol errors
 Higher bandwidth => higher symbol rate, and smaller time per-
symbol
 Lower symbol rate, more time, energy per-symbol
 Multipaths & Bandwidth
31

 Smaller bandwidth => fewer channel

taps (OFDM)
 Larger bandwidth In CDMA: Rake
(Equalization) Receiver is used, the
power on multipath taps is received
(“rake fingers”), gain adjusted and
combined.
SLOW VS FAST FADING
 Doppler: Time Variations in Model
33

 Due to the presence of motion at the

transmitter, the receiver or the
surrounding objects, the multipath
channel changes over time.
 Channel coherence time: time range
over which the correlation between the
average power associated with a given
multipath delay at two different time
instants is approximately zero. Tc
depends on carrier frequency and
vehicular speed
 Time-Varying Channel Impulse Response
34
 Doppler: Time Variations in Model
35

 Doppler spread is the inverse of Channel coherence time.

 Slow fading channel:
symbol period < channel coherence time .
signal bandwidth > Doppler spread.
 Fast fading channel:
symbol period > channel coherence time .
signal bandwidth < Doppler spread .
FADING SUMMARY
 Fading Summary
37
 Fading Summary
38

Where 𝑇𝑠 is the symbol duration, ; 𝑇𝐶 is the channel coherence time, and 𝜎𝜏 is the
channel delay spread
 Fading Summary
39

Where 𝐵𝑠 is the symbol bandwidth, ; 𝐵𝐶 is the channel coherence BW, and 𝐵𝑑 is the
channel Doppler spread
STATISTICAL CHANNEL MODEL
 Statistical channel model
41

Discrete-time equivalent baseband

model of the channel:

Frequency selective

Flat fading
 Clarcks model:
42
 •According to the central limit theorem, if the
number of such paths are large enough then the
amplitudes follow the complex Gaussian
distribution
 Fading with many scatterers: Central Limit Theorem
 In-phase (cosine) and quadrature (sine)
components of the r(0), denoted as rI (0) and rQ(0)
are independent Gaussian random variables.

 Envelope Amplitude: 𝑟 = 𝑟𝐼2 + 𝑟𝑄2 is Rayleigh.

 phase arg(ℎ) follows a uniform distribution
over [0,2𝜋].
 Rayleigh, Ricean fading
43

• A few major multipaths, and lots of local scatterers

=> each channel sample “tap” can be modeled as
Rayleigh

• Ricean distribution is used when there is a dominant LOS

path plus scattered paths
• K parameter: strength of LOS to non-LOS.
• K = 0 => Rayleigh
 Clarcks model:
44

 ℎ𝑘 [𝑚] is modeled as circular symmetric processes

independent across the taps.
2
 If for all taps, ℎ𝑘 [𝑚]~𝐶𝑁 0, 𝜎𝑙
the model is called Rayleigh.
 If for one tap, 1 line-of-sight plus scattered paths

the model is called Rician with K-factor .

 Effect of Rayleigh Fading
45

Solution: use channel coding to

improve the BER

Required using Matlab

The End