Lecture 20

Intersymbol Interference (3.3 complete article)

Equalization 3.4.1, 3.4.2 only
 DCS MODEL
 Baseband system model:
x1 x2
xk  Tx filter Channel r (t ) Rx. filter
zk
x̂k 
ht (t ) hc (t ) hr (t ) Detector
t  kT
T Ht ( f ) Hc ( f ) Hr ( f )
x3 T n(t )
 Equivalent model:
H
( 
f) H
t(
f)H(
c f)
H (
r f)
x1 x2
Equivalent system
xk  z (t ) zk
x̂k 
h(t ) Detector
t  kT
T H( f )
x3 T nˆ (t )
filtered noise
 When the receiving filter is configured to compensate for the
distortion caused by both the transmitter and the channel, it is often
1
referred to as an equalizing filter 𝐻𝑒 𝑓 =
𝐻𝑡 𝑓 𝐻𝑐 (𝑓)
2
 3.3 INTER-SYMBOL INTERFERENCE (ISI)
 Input data stream and bit superposition

 The channel output is the sum of the contributions from each bit

3
 ISI occurs whenever a non-bandlimited line code is used over a
bandlimited channel
 ISI effects at the sampling instants
 Overlapping pulses will not cause ISI if they have zero amplitude at
the time the signal is sampled

4
NYQUIST FILTERS
 Nyquist bandwidth condition: The theoretical minimum required
system bandwidth to detect 𝑅𝑠 (symbols/s) without ISI is 𝑅𝑠/2 (Hz).
 This occurs when the system transfer function 𝐻(𝑓 ) is made
rectangular. For baseband systems, when 𝐻(𝑓) is such a filter with
single-sided bandwidth 1/2𝑇 , it is known as the ideal Nyquist filter.
 If each pulse of the received signal is of the form sinc(𝑡 𝑇), the pulses
can be detected without ISI. This sinc shaped pulse is called the ideal
Nyquist pulse.

5
IDEAL NYQUIST PULSE (FILTER)
Ideal Nyquist filter Ideal Nyquist pulse

H( f ) 1
W
2T (t)sinc(
h t/T)
T 1

0 f  2T  T 0 T 2T t
1 1
2T 2T

6
 BANDWIDTH EFFICIENCY
 Bandwidth efficiency is an important measure in DCS representing
data throughput per hertz of bandwidth 𝑅/𝑊 (bits/s/Hz)
 Shows how efficiently the bandwidth resources are used by signaling
techniques.
 For baseband systems, bandwidth required to detect 1 ∕ 𝑇 symbols/sec
is 1 2𝑇 Hz
1 R R
 s  W  s  2 (symbol/s/Hz)
2T 2 W

 A system with bandwidth 𝑊 = 1/2𝑇 = 𝑅𝑠/2 (Hz) can support a

maximum transmission rate of 2𝑊 = 1/𝑇 = 𝑅𝑠 (symbols/s) without
ISI.→ Nyquist Bandwidth constraint
 For ideal Nyquist filtering and zero ISI, the maximum possible symbol
transmission rate per Hz is 2 symbols/s/Hz →symbol rate packing
7
 THE RAISED COSINE FILTER
 The raised-cosine filter is a filter frequently used for pulse-shaping in
digital modulation due to its ability to minimize ISI.
 Its name stems from the fact that the non-zero portion of the frequency
spectrum of its simplest form is a cosine function, 'raised' up to sit
above the f axis.
 The Nyquist Class (Zero ISI at the sampling time)
1 for
|f| 2
W W

0

2|f|
W2
W 
H
(
f)

cos 0
for
2W W 
|f| W

0
 4 W W0 

0 for
|f| W

h
(
t
)2
W(sinc(
2
Wt
))

cos[
2(WW
0)
t] 
 
0 0 2
1[
4
(WW)
t
0]
where 𝑊 is the absolute bandwidth and 𝑊0 = 1/2𝑇 represents the
minimum Nyquist bandwidth for the rectangular spectrum and the –6 dB
bandwidth (or half-amplitude point) for the raised-cosine spectrum
Excess bandwidth: W  W0 WW0 8
Roll-off factor r  0  r 1
W
THE RAISED COSINE FILTER – CONT’D

(f)|
|H |H (f)|
RC
h(t)hRC(t)

1 r 0 1

r  0.5 r 1
0.5 0.5
r 1 r  0.5
r 0

 2T T 0 T 2T 3T
1  3 1 0 1 3 1  3T
T 4T 2T 2T 4T T

Larger the filter roll-off , the shorter will be the pulse tail, and sensitivity to timing errors
Smaller the filter roll-off, greater the signaling rate / no of users
General expression for requires BW and symbol rate
Rs
Baseband W SSB  (1  r ) W 
(
Passband
1r)
R
2 DSB s
9
 3.3.1.2 ROOT-RAISED COSINE FILTER
 The raised-cosine frequency transfer function describes the
composite (full round trip) form inception of message (as an
impulse) at the transmitter, through the channel and through the
receiving filter.
 The filtering at the receiver compensates for the overall transfer
function to bring zero ISI,
 Raised cosine filter is one such overall transfer function to bring zero
ISI. Often this is accomplished by choosing (matching) the receiving
filter and the transmitting filter so that each has a transfer function
known as root-raised cosine (square root of raised cosine).
 Neglecting any channel-induced ISI, the product of these root-raised
cosine functions yields the composite raised-cosine system transfer
function.

10
SQUARE-ROOT NYQUIST PULSE

11
Article 3.3.3.2
SUMMARY: PULSE SHAPING TO REDUCE ISI
 Data transmission over bandlimited channels requires pulse
shaping to eliminate or control ISI.
 Nyquist filters provide ISI-free transmission

 Goals and trade-off in pulse-shaping

- Reduce ISI
- Efficient bandwidth utilization
- Robustness to timing error (small side lobes)

12
3.3.2 TYPES OF ERROR-PERFORMANCE DEGRADATION

13
PULSE SHAPING AND EQUALIZATION TO REMOVE ISI
 To make the signal free from ISI, and to ensure a maximum signal to
noise ratio, we need to implement a method called Equalization.
 Whenever a separate equalizing filter is introduced to mitigate the
effects of channel-induced ISI, the receiving and equalizing filters
together should be configured to compensate for the distortion
caused by both the transmitter and the channel so as to yield an
overall system transfer function characterized by zero ISI.
No ISI at the sampling time
H(
f
RC)
H(
t f
)H(
cf)
Hr(
f)
He(
f)

 Square-Root Raised Cosine (SRRC) filter and Equalizer

H
(f
)
RCH
(
t f
)H(
rf)
Taking care of ISI
H
(
f
)H
r(
f
) H
(f
)H(
f)caused by tr. filter
t RC SRRC

1
e(f)
H Taking care of ISI
H caused by channel
c(f) 14
EQUALIZATION
 ISI due to filtering effect of the communications channel (e.g.
wireless channels)
 Channels behave like band-limited filters


H
c(f)Hc(f)e j c(f)

Non-constant amplitude Non-linear phase

Amplitude distortion Phase distortion

 When the receiving filter is configured to compensate for the

distortion caused by both the transmitter and the channel, it is often
referred to as an equalizing filter.
15
 EQUALIZATION – CONT’D

Step 1 – waveform to sample transformation Step 2 – decision making

Demodulate & Sample Detect

z (T ) m̂i
Threshold
r (t ) Frequency Receiving Equalizing
comparison
down-conversion filter filter

Compensation
For bandpass signals
for channel
induced ISI

Received waveform Baseband pulse

Baseband pulse Sample
(possibly distored)
(test
statistic)

16
 3.4.2 EYE PATTERN
 Display on an oscilloscope which sweeps the system response to a
baseband signal at the rate 1/T (T symbol duration)

Distortion
due to ISI 𝐷𝐴
amplitude scale

Noise margin 𝑀𝑁

Sensitivity to
timing error 𝑆𝑇

Timing jitter 𝐽𝑇
time scale 17
 The width of the eye opening indicates the time over which
sampling for detection might be performed. The optimum
sampling time corresponds to the maximum eye opening, yielding
the greatest protection against noise.
 The range of amplitude differences labeled 𝐷𝐴 is a measure of
distortion caused by ISI,
 The range of time differences of the zero crossings labeled 𝐽𝑇 is a
measure of the timing jitter.
 Measures of noise margin 𝑀𝑁 and sensitivity-to-timing error 𝑆𝑇 are
also shown in the figure.
 In general, the most frequent use of the eye pattern is for
qualitatively assessing the extent of the ISI. As the eye closes, ISI is
increasing; as the eye opens, ISI is decreasing.

18