Scribd
Upload
114 views13 pages

Channel Sounding

Chapter_8

Uploaded by

Walid Mohanna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
114 views13 pages

Channel Sounding

Chapter_8

Uploaded by

Walid Mohanna
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 13

Chapter 8

Channel sounding
Stochastic Channel Models - parameter values
have to be obtained from measurements
Deterministic Channel Models - quality of the
prediction has to be checked by comparisons
with measured data

Slides for Wireless Communications Edfors, Molisch, Tufvesson

195

Channel measurements
In order to model the channel behavior we need to measure
for wideband systems we'll measure the delay dispersion (impluse response)
its properties versus just measuring the signal strength as was done for narrowband 2G systems
Time domain measurements
impulse sounder
correlative sounder

Frequency domain measurements


Vector network analyzer

Directional measurements

directional antennas
real antenna arrays
multiplexed arrays requires double-directional impulse responses
virtual arrays
Slides for Wireless Communications Edfors, Molisch, Tufvesson

196

Channel Sounder Block Diagram


s(t) periodic pulses

Transmitter and receiver are closely synchronized (cable or GPS)


Duration of the sounding signal not longer than channel coherence time (constant channel)
Time bandwidth of the sounding signal is longer than the inverse of the system bandwidth

Constant power-spectral density uniform across the signal bandwidth with no side lobes
Low crest factor where Ccrest = Peak Amplitude / RMS Amplitude which makes efficient
use of transmitter power/dynamic range

Basic identifiability of the channel


The channel can be measured uniquely only if the following
a time-variant channel where the
two opposing conditions are satisfied Mobile
Station (MS) can be moving
This is a sampling theorem in the time
domain - a sampling rate to identify a
time-variant process with a band-limited
Doppler spectrum

f rep 2 max

Trep =

1
f rep

max

the sampling frequency must be twice the maximum


Doppler shift frequency Vmax = velocity/wavelength

the pulse time (width) must be more than the max excess delay
of the channel or the pulses overlap (time for the delays to die out)
which means the pulse time is less than the channel's coherence time
(channel constant period/time invariant).

Therefore, a channel can only be measured uniquely if it is

underspread

a compromise of sorts since the two limitations above are counter to each other

2max max 1

See Example 8.1 Page 149 where Trep = 7.5 mS


which is a much larger period than the normal wireless
channel delays which are in the nanoSecond (nS) range

This condition is fulfilled in all practical wireless applications


fulfills Nyquist Criterion and implies a slowly time variant channel which has a long coherence time

Slides for Wireless Communications Edfors, Molisch, Tufvesson

197

Impulse sounder

h meas t i , p ht i ,

Assuming a slowly time-variant channel.


The impulse response of the sounder is the
convolution of the TX pulse shape with the
RX filter impulse response which is a
bandpass filter thus the receiver should not
have an impact on h.

impulse response
of sounder
impulse response
of channel
~

Slides for Wireless Communications Edfors, Molisch, Tufvesson

198

Correlative sounder
Other signals on channel (interference) make measurements difficult although
the interference can usually be approximated as equivalent Gaussian noise.
Concatenation of the transmit and the receive filters have an impulse response that is identical to the autocorrelation (ACF) of the transmit
filter resulting in an impulse response that is a good approximattion of a delta function. In practice, PseudoNoise (PN) sequences or
chirp signals (linear frequency modulated signals) have become the prevalent sounding sequences for correlative sounder techniques.

p()

h( )

-Tc

h( )

Tc

correlation peak

impulse response

measured impulse response

Slides for Wireless Communications Edfors, Molisch, Tufvesson

199

Frequency domain measurements


Use a vector network analyzer or similar to determine the
transfer function of the channel

H meas ( f ) HTXantenna ( f )* Hchannel ( f )* H RXantenna ( f )


Need to know the influence of the measurement system
In practice Pseudo Noise (PN) sequences have become the prevalent sounding
sequences where maximum-length PN sequences (m-sequences used in CDMA
systems) are most popular.
The principles behind OFDM to transmit multiple frequencies at the same time
(IFFT Chapter 19) are used to measure the transfer function of the channel
in the frequency domain.

Slides for Wireless Communications Edfors, Molisch, Tufvesson

200

Channel sounding directional antenna


for MIMO

Measure one impulse


response for each
antenna orientation

Slides for Wireless Communications Edfors, Molisch, Tufvesson

201

Channel sounding antenna array

Measure one impulse


response for each antenna
element
Ambiguity with linear array
d

h()

h()

x=0

linear array

h()

x=d

h()

x=2d

x=(M-1)d

Signal processing
spatially resolved impulse response

Slides for Wireless Communications Edfors, Molisch, Tufvesson

202

Real, multiplexed, and virtual arrays


Real array: simultaneous
measurement at all
antenna elements
Multiplexed array: short
time intervals between
measurements at different
elements

RX

RX

RX

Digital Signal Processing

RX

Digital Signal Processing

Virtual array: long delay


no problem with mutual
coupling

mechanical
movement
of antenna
RX

Digital Signal Processing

Slides for Wireless Communications Edfors, Molisch, Tufvesson

Copyright: Hindawi

203

Directional analysis

The DoA can, e.g., be


estimated by correlating the
received signals with steering
vectors.
1
expjk 0 d cos

expj2k 0 d cos

expjM 1k 0 d cos

d sin

An element spacing of d=5.8 cm


and an angle of arrival of =20
degrees gives a time delay of
6.610-11 s between neighboring
elements

Slides for Wireless Communications Edfors, Molisch, Tufvesson

204

High resolution algorithms


In order to get better angular resolution, other techniques
for estimating the angles are used, e.g.:

MUSIC, subspace method using spectral search


ESPRIT, subspace method
MVM (Capons beamformer), rather easy spectral search method
SAGE, iterative maximum likelihood method

Based on models for the propagation


Rather complex, one measurement point may take 15
minutes on a decent computer

Slides for Wireless Communications Edfors, Molisch, Tufvesson

205

Antenna array TX
Transmission must be done so
that RX can distinguish signals
from different TX antennas
Transmit signals should be
orthogonal
tra ns mit
signal

Orthogonality in time
time t

< inverse Do ppler frequency

Orthogonality in frequency

tra ns mit
signal

Antenna 1
fre quenc y f

< c ohe renc e ba ndwid th

Orthogonality in code

Antenna 2
Antenna 3

tra ns mit
signal

time t

Copyright: Hindawi

Slides for Wireless Communications Edfors, Molisch, Tufvesson

206

You might also like