You are on page 1of 32

9/20/2013 1

Digital communications
Chapter 4:
Impairment source - ISI
Referenced: C. Logothetis (se), Comlab.hut.fin,
Forouzan Mc.Graw Hill
Lectured by Assoc. Prof. Thuong Le-Tien
Septermber 2013
9/20/2013 2
Sources of error:
AWGN
Inter-symbol interference (ISI)
Nyquist theorem
The techniques to reduce ISI
Pulse shaping
Equalization
9/20/2013 3
Inter-Symbol Interference (ISI)
ISI in the detection process due to the
filtering effects of the system
Overall equivalent system transfer function
creates echoes and hence time dispersion
causes ISI at sampling time
) ( ) ( ) ( ) ( f H f H f H f H
r c t
=
i
k i
i k k k
s n s z

=
+ + = o
9/20/2013 4
Baseband system model
Equivalent model
Tx filter
Channel
) (t n
) (t r
Rx. filter
Detector
k
z
kT t =
{ }
k
x { }
k
x
1
x
2
x
3
x
T
T
) (
) (
f H
t h
t
t
) (
) (
f H
t h
r
r
) (
) (
f H
t h
c
c
Equivalent system
) ( t n
) (t z
Detector
k
z
kT t =
{ }
k
x { }
k
x
1
x
2
x
3
x
T
T
) (
) (
f H
t h
filtered noise
) ( ) ( ) ( ) ( f H f H f H f H
r c t
=
9/20/2013 5
Nyquist bandwidth constraint
Nyquist bandwidth constraint:
The theoretical minimum required system bandwidth to
detect R
s
[symbols/s] without ISI is R
s
/2 [Hz].
Equivalently, a system with bandwidth W=1/2T=R
s
/2
[Hz] can support a maximum transmission rate of
2W=1/T=R
s
[symbols/s] without ISI.
Bandwidth efficiency, R/W [bits/s/Hz] :
An important measure in DCs representing data
throughput per hertz of bandwidth.
Showing how efficiently the bandwidth resources are
used by signaling techniques.
Hz] [symbol/s/ 2
2 2
1
> s =
W
R
W
R
T
s s
9/20/2013 6
Ideal Nyquist pulse (filter)
T 2
1
T 2
1
T
) ( f H
f t
) / sinc( ) ( T t t h =
1
0
T T 2 T T 2
0
T
W
2
1
=
Ideal Nyquist filter Ideal Nyquist pulse
9/20/2013 7
Nyquist pulses (filters)
Nyquist pulses (filters):
Pulses (filters) which results in no ISI at the
sampling time.
Nyquist filter:
Its transfer function in frequency domain is
obtained by convolving a rectangular function with
any real even-symmetric frequency function
Nyquist pulse:
Its shape can be represented by a sinc(t/T)
function multiply by another time function.
Example of Nyquist filters: Raised-Cosine filter
9/20/2013 8
How can ISI can be reduced?
Goals and trade-off in pulse-shaping
Reduce ISI
Efficient bandwidth utilization
Robustness to timing error (small side
lobes)
9/20/2013 9
The raised cosine filter
A Nyquist pulse (No ISI at the sampling time)

>
< <
(

+
<
=
W f
W f W W
W W
W W f
W W f
f H
| | for 0
| | 2 for
2 | |
4
cos
2 | | for 1
) (
0
0
0
2
0
t
Excess bandwidth:
0
W W
Roll-off factor
0
0
W
W W
r

=
1 0 s s r
2
0
0
0 0
] ) ( 4 [ 1
] ) ( 2 cos[
)) 2 (sinc( 2 ) (
t W W
t W W
t W W t h


=
t
9/20/2013 10
2
) 1 ( Baseband
sSB
s
R
r W + =
| ) ( | | ) ( | f H f H
RC
=
0 = r
5 . 0 = r
1 = r
1 = r
5 . 0 = r
0 = r
) ( ) ( t h t h
RC
=
T 2
1
T 4
3
T
1
T 4
3
T 2
1
T
1
1
0.5
0
1
0.5
0 T T 2 T 3 T T 2
T 3
s
R r W ) 1 ( Passband
DSB
+ =
9/20/2013 11
Pulse shaping and equalization to remove ISI
Square-Root Raised Cosine (SRRC) filter and Equalizer
) ( ) ( ) ( ) ( ) (
RC
f H f H f H f H f H
e r c t
=
No ISI at the sampling time
) ( ) ( ) ( ) (
) ( ) ( ) (
SRRC RC
RC
f H f H f H f H
f H f H f H
t r
r t
= = =
=
Taking care of ISI
caused by transmittion
filter
) (
1
) (
f H
f H
c
e
=
Taking care of ISI
caused by channel
9/20/2013 12
Example of pulse shaping
Square-root Raised-Cosine (SRRC) pulse shaping
t/T
Amp. [V]
Baseband tr. Waveform
Data symbol
First pulse
Second pulse
Third pulse
9/20/2013 13
Raised Cosine pulse at the output of matched filter
t/T
Amp. [V]
Baseband received waveform at
the matched filter output
(zero ISI)
9/20/2013 14
Eye pattern
Eye pattern:Display on an oscilloscope which
sweeps the system response to a baseband signal at
the rate 1/T (T symbol duration)
time scale
a
m
p
l
i
t
u
d
e

s
c
a
l
e
Noise margin
Sensitivity to
timing error
Distortion
due to ISI
Timing jitter
9/20/2013 15
Example of eye pattern:
Perfect channel (no noise and no ISI)
9/20/2013 16
AWGN (Eb/N0=20 dB) and no ISI
9/20/2013 17
AWGN (Eb/N0=10 dB) and no ISI
9/20/2013 18
Equalization
Frequency
down-conversion
Receiving
filter
Equalizing
filter
Threshold
comparison
For bandpass signals
Compensation for
channel induced ISI
Baseband pulse
(possibly distored)
Sample
(test statistic)
Baseband pulse
Received waveform
Step 1 waveform to sample transformation Step 2 decision making
) (t r
) (T z
i
m
Demodulate & Sample Detect
9/20/2013 19
Equalization
ISI due to filtering effect of the
communications channel (e.g. wireless
channels)
Channels behave like band-limited filters
) (
) ( ) (
f j
c c
c
e f H f H
u
=
Non-constant amplitude
Amplitude distortion
Non-linear phase
Phase distortion
9/20/2013 20
Equalization: Channel examples
Example of a frequency selective, slowly changing (slow fading)
channel for a user at 35 km/h
9/20/2013 21
Equalization: Channel examples
Example of a frequency selective, fast changing (fast fading)
channel for a user at 35 km/h
9/20/2013 22
Example of eye pattern with ISI:
Binary-PAM
Non-ideal channel and no noise
) ( 7 . 0 ) ( ) ( T t t t h
c
+ = o o
9/20/2013 23
Example of eye pattern with ISI:
Binary-PAM
AWGN (Eb/N0=20 dB) and ISI
) ( 7 . 0 ) ( ) ( T t t t h
c
+ = o o
9/20/2013 24
Equalizing filters
Baseband system model
Equivalent model
Tx filter
Channel
) (t n
) (t r
Rx. filter
Detector
k
z
kT t =
{ }
k
a
1
a
2
a
3
a
T
) (
) (
f H
t h
t
t
) (
) (
f H
t h
r
r
) (
) (
f H
t h
c
c
Equivalent system
) ( t n
) (t z
Detector
k
z
kT t =
) (
) (
f H
t h
filtered noise
) ( ) ( ) ( ) ( f H f H f H f H
r c t
=


k
k
kT t a ) ( o
Equalizer
) (
) (
f H
t h
e
e
1
a
2
a
3
a
T


k
k
kT t a ) ( o
) (t x
Equalizer
) (
) (
f H
t h
e
e
) ( ) ( ) ( t h t n t n
r
- =
{ }
k
a ) (t z
) (t z
9/20/2013 25
Equalization using
MLSE (Maximum likelihood sequence
estimation)
Filtering
Transversal filtering
Zero-forcing equalizer
Minimum mean square error (MSE) equalizer
Decision feedback
Using the past decisions to remove the ISI contributed
by them
Adaptive equalizer
9/20/2013 26
Equalization by transversal filtering
Transversal filter:
A weighted tap delayed line that reduces the effect
of ISI by proper adjustment of the filter taps.

=
= = =
N
N n
n
N N k N N n n t x c t z 2 ,..., 2 ,..., ) ( ) ( t
t t t t
N
c
1 + N
c
1 N
c
N
c

) (t x
) (t z
Coeff.
adjustment
9/20/2013 27
Zero-forcing equalizer:
The filter taps are adjusted such that the equalizer output
is forced to be zero at N sample points on each side:
Mean Square Error (MSE) equalizer:
The filter taps are adjusted such that the MSE of ISI and
noise power at the equalizer output is minimized.
N k
k
k z
=
=

=
,..., 1
0
0
1
) (
{ }
N
N n
n
c
=
Adjust
| |
2
) ) ( ( min
k
a kT z E
{ }
N
N n
n
c
=
Adjust
9/20/2013 28
Example of equalizer
2-PAM with SRRQ
Non-ideal channel
One-tap DFE
) ( 3 . 0 ) ( ) ( T t t t h
c
+ = o o
Matched filter outputs at the sampling time
ISI-no noise,
No equalizer
ISI-no noise,
DFE equalizer
ISI- noise
No equalizer
ISI- noise
DFE equalizer
9/20/2013 29
Equalization: Removing Residual ISI
Consider a tapped delay line equalizer with
Search for the tap gains c
N
such that the output equals zero at
sample intervals D except at the decision instant when it should
be unity. The output is (think for instance paths c
-N
, c
N
or c
0
)
that is sampled at yielding
( ) ( )
=

=
N
eq n
n N
p t c p t nD ND
| |
( ) ( ) ( )
N N
eq n n
n N n N
p kD ND c p kD nD c p D k n
= =

+ = =
= +
k
t kD ND
( ) ( 2 )
N N
p t c p t ND =
( ) ( )
N N
p t c p t

=
0
( ) ( )
N
p t c p t ND

=
9/20/2013 30
Tapped Delay Line:
Matrix Representation
At the instant of decision:
That leads into (2N+1)x(2N+1) matrix where (2N+1) tap
coefficients can be solved:
| |
1, 0
( ) ( )
0, 1, 2,...,
N N
eq k n n
n N n N k n
k
p t c p D k n c p
k N
= =

=

= = =

=


0 2
1 1 1
0
1 1 1
2 0
... 0
...
... 0
... 1
... 0
... 0

+ +
( ( (
( ( (
( ( (
( ( (
( ( (
=
( ( (
( ( (
( ( (
( ( (
( ( (








N N
N N
N N
N N
N N
p p c
p p c
p p c
p p c
p p c
0 1 1 2
1 0 1 2 1
1 1
2 2 1 1 0
... 0
... 0
... 1
... 0
n n n n
n n n n
n n n n n n
n n n n n
p c p c p c
p c p c p c
p c p c p c
p c p c p c
+
+ +
+
+
+ + + =
+ + + =
+ + + =
+ + + =


0 1 2
1
n
n n n
p p p
c c c

+

+
( )
eq
p t
( ) p t
9/20/2013 31
Example of Equalization
Read the distorted pulse values into
matrix from fig. (a)
and the solution is
1
0
1
1.0 0.1 0.0 0
0.2 1.0 0.1 1
0.1 0.2 1.0 0

( ( (
( ( (
=
( ( (
( ( (

c
c
c
1
0
1
0.096
0.96
0.2


( (
( (
=
( (
( (

c
c
c
Zero forced values
0
p
1
p
2
p
1
p

2
p

Question: what does these


zeros help because they dont
exist at the sampling instant?
9/20/2013 32
Monitoring Transmission Quality
by Eye Diagram
Required minimum bandwidth is
Nyqvists sampling theorem:
/ 2
T
B r >
Given an ideal LPF with the
bandwidth B it is possible to
transmit independent
symbols at the rate:
/ 2 1/(2 )
T b
B r T > =

You might also like