P. 1
HW#6

|Views: 2|Likes:

See more
See less

05/04/2012

pdf

text

original

# 14

the bit rate is 5334 bps.
(c) M = 8 QAM with a pulse shape that has β =
1
2
. From (a), the symbol rate is
1
T
= 2667 and
hence the bit rate
3
T
= 8001 bps.
(d) Binary FSK with noncoherent detection. Assuming that the frequency separation between the
two frequencies is ∆f =
1
T
, where
1
T
is the bit rate, the two frequencies are f
c
+
1
2T
and f
c

1
2T
.
Since W = 4000 Hz, we may select
1
2T
= 1000, or, equivalently,
1
T
= 2000. Hence, the bit rate is
2000 bps, and the two FSK signals are orthogonal.
(e) Four FSK with noncoherent detection. In this case we need four frequencies with separation
of
1
T
between adjacent frequencies. We select f
1
= f
c

1.5
T
, f
2
= f
c

1
2T
, f
3
= f
c
+
1
2T
, and
f
4
= f
c
+
1.5
T
, where
1
2T
= 500 Hz. Hence, the symbol rate is
1
T
= 1000 symbols per second and
since each symbol carries two bits of information, the bit rate is 2000 bps.
(f) M = 8 FSK with noncoherent detection. In this case we require eight frequencies with frequency
separation of
1
T
= 500 Hz for orthogonality. Since each symbol carries 3 bits of information, the
bit rate is 1500 bps.
Problem 9.13
(a) The bandwidth of the bandpass channel is :
W = 3000 −600 = 2400 Hz
Since each symbol of the QPSK constellation conveys 2 bits of information, the symbol rate of
transmission is :
R =
1
T
=
2400
2
= 1200 symbols/sec
Thus, for spectral shaping we can use a signal pulse with a raised cosine spectrum and roll-oﬀ factor
β = 1, since the spectral requirements will be
1
2T
(1 +β) =
1
T
= 1200Hz. Hence :
X
rc
(f) =
T
2
[1 + cos(πT|f|)] =
1
1200
cos
2

π|f|
2400

If the desired spectral characteristic is split evenly between the transmitting ﬁlter G
T
(f) and the
receiving ﬁlter G
R
(f), then
G
T
(f) = G
R
(f) =

1
1200
cos

π|f|
2400

, |f| <
1
T
= 1200
A block diagram of the transmitter is shown in the next ﬁgure.
l
×
E E
T
E to Channel
cos(2πf
c
t)
G
T
(f)
QPSK
a
n
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
15
(b) If the bit rate is 4800 bps, then the symbol rate is
R =
4800
2
= 2400 symbols/sec
In order to satisfy the Nyquist criterion, the the signal pulse used for spectral shaping, should have
roll-oﬀ factor β = 0 with corresponding spectrum :
X(f) = T, |f| < 1200
Thus, the frequency response of the transmitting ﬁlter is G
T
(f) =

T, |f| < 1200.
Problem 9.14
The bandwidth of the bandpass channel is :
W = 3300 −300 = 3000 Hz
In order to transmit 9600 bps with a symbor rate R =
1
T
= 2400 symbols per second, the number
of information bits per symbol should be :
k =
9600
2400
= 4
Hence, a 2
4
= 16 QAM signal constellation is needed. The carrier frequency f
c
is set to 1800 Hz,
which is the mid-frequency of the frequency band that the bandpass channel occupies. If a pulse
with raised cosine spectrum and roll-oﬀ factor β is used for spectral shaping, then for the bandpass
signal with bandwidth W :
1
2T
(1 +β) =
W
2
= 1500 ⇒β = 0.25
A sketch of the spectrum of the transmitted signal pulse is shown in the next ﬁgure.
-1800 -300 -3300 300 3300 1800
1/2T
f
900
2700
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
29
The term I
m−1
A
2
T
10
expresses the ISI introduced to the system. If I
m
= 1 is transmitted, then the
probability of error is
P(e|I
m
= 1) =
1
2
P(e|I
m
= 1, I
m−1
= 1) +
1
2
P(e|I
m
= 1, I
m−1
= −1)
=
1
2

πN
0
A
2
T
_
−A
2
T
−∞
e

ν
2
N
0
A
2
T
dν +
1
2

πN
0
A
2
T
_

8
10
A
2
T
−∞
e

ν
2
N
0
A
2
T

=
1
2
Q
_
_
¸
2A
2
T
N
0
_
_
+
1
2
Q
_
_
¸
_
8
10
_
2
2A
2
T
N
0
_
_
Since the symbols of the binary PAM system are equiprobable the previous derived expression is
the probability of error when a symbol by symbol detector is employed. Comparing this with the
probability of error of a system with no ISI, we observe that there is an increase of the probability
of error by
P
diﬀ
(e) =
1
2
Q
_
_
¸
_
8
10
_
2
2A
2
T
N
0
_
_

1
2
Q
_
_
¸
2A
2
T
N
0
_
_
Problem 9.34
(a) Taking the inverse Fourier transform of H(f), we obtain :
h(t) = F
−1
[H(f)] = δ(t) +
α
2
δ(t −t
0
) +
α
2
δ(t +t
0
)
Hence,
y(t) = s(t) ⋆ h(t) = s(t) +
α
2
s(t −t
0
) +
α
2
s(t +t
0
)
(b) If the signal s(t) is used to modulate the sequence {I
n
}, then the transmitted signal is :
u(t) =

n=−∞
I
n
s(t −nT)
The received signal is the convolution of u(t) with h(t). Hence,
y(t) = u(t) ⋆ h(t) =
_

n=−∞
I
n
s(t −nT)
_

_
δ(t) +
α
2
δ(t −t
0
) +
α
2
δ(t +t
0
)
_
=

n=−∞
I
n
s(t −nT) +
α
2

n=−∞
I
n
s(t −t
0
−nT) +
α
2

n=−∞
I
n
s(t +t
0
−nT)
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
30
Thus, the output of the matched ﬁlter s(−t) at the time instant t
1
is :
w(t
1
) =

n=−∞
I
n
_

−∞
s(τ −nT)s(τ −t
1
)dτ
+
α
2

n=−∞
I
n
_

−∞
s(τ −t
0
−nT)s(τ −t
1
)dτ
+
α
2

n=−∞
I
n
_

−∞
s(τ +t
0
−nT)s(τ −t
1
)dτ
If we denote the signal s(t) ⋆ s(t) by x(t), then the output of the matched ﬁlter at t
1
= kT is :
w(kT) =

n=−∞
I
n
x(kT −nT)
+
α
2

n=−∞
I
n
x(kT −t
0
−nT) +
α
2

n=−∞
I
n
x(kT +t
0
−nT)
(c) With t
0
= T and k = n in the previous equation, we obtain :
w
k
= I
k
x
0
+

n=k
I
n
x
k−n
+
α
2
I
k
x
−1
+
α
2

n=k
I
n
x
k−n−1
+
α
2
I
k
x
1
+
α
2

n=k
I
n
x
k−n+1
= I
k
_
x
0
+
α
2
x
−1
+
α
2
x
1
_
+

n=k
I
n
_
x
k−n
+
α
2
x
k−n−1
+
α
2
x
k−n+1
_
The terms under the summation is the ISI introduced by the channel. If the signal s(t) is designed
so as to satisfy the Nyquist criterion, then :
x
k
= 0, k = 0
and the aobove expression simpliﬁes to :
w
k
= I
k
+
α
2
(I
k+1
+I
k−1
)
Problem 9.35
(a) Each segment of the wire-line can be considered as a bandpass ﬁlter with bandwidth W = 1200
Hz. Thus, the highest bit rate that can be transmitted without ISI by means of binary PAM is :
R = 2W = 2400 bps
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
35
Since the information symbols are uncorrelated : E [I
k
I

l
] = δ
kl
. We also have :
E [I
k
u

l
] = E

I
k

L
m=0
f

m
I

l−m
+ n

l

= f

l−k
Hence, equation (1) gives :
E

I
k
I

k−l

= E

ˆ
I
k
I

k−l

, 1 ≤ l ≤ K
2

0 = E

0
j=−K
1
c
j
u
k−j
+

K
2
j=1
c
j
I
k−j

I

k−l

0 =

0
j=−K
1
c
j
f
l−j

+ c
l

c
l
= −

0
j=−K
1
c
j
f
l−j
, 1 ≤ l ≤ K
2
which is the desired equation for the feedback taps.
Problem 9.41
(a) The equivalent discrete-time impulse response of the channel is :
h(t) =
1

n=−1
h
n
δ(t −nT) = 0.3δ(t + T) + 0.9δ(t) + 0.3δ(t −T)
If by {c
n
} we denote the coeﬃcients of the FIR equalizer, then the equalized signal is :
q
m
=
1

n=−1
c
n
h
m−n
which in matrix notation is written as :

0.9 0.3 0.
0.3 0.9 0.3
0. 0.3 0.9

c
−1
c
0
c
1

=

0
1
0

The coeﬃcients of the zero-force equalizer can be found by solving the previous matrix equation.
Thus,

c
−1
c
0
c
1

=

−0.4762
1.4286
−0.4762

PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
36
(b) The values of q
m
for m = ±2, ±3 are given by
q
2
=
1

n=−1
c
n
h
2−n
= c
1
h
1
= −0.1429
q
−2
=
1

n=−1
c
n
h
−2−n
= c
−1
h
−1
= −0.1429
q
3
=
1

n=−1
c
n
h
3−n
= 0
q
−3
=
1

n=−1
c
n
h
−3−n
= 0
Problem 9.42
(a) The output of the zero-force equalizer is :
q
m
=
1

n=−1
c
n
x
mn
With q
0
= 1 and q
m
= 0 for m = 0, we obtain the system :

1.0 0.1 −0.5
−0.2 1.0 0.1
0.05 −0.2 1.0

c
−1
c
0
c
1

=

0
1
0

Solving the previous system in terms of the equalizer’s coeﬃcients, we obtain :

c
−1
c
0
c
1

=

0.000
0.980
0.196

PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
39
and therefore :
P
2
= Q
_
_
8E
b
4.5958N
0
_
The bit error probability is
P
2
2
.
Problem 9.44
The optimum tap coeﬃcients of the zero-force equalizer can be found by solving the system:
_
_
_
_
1.0 0.3 0.0
0.2 1.0 0.3
0.0 0.2 1.0
_
_
_
_
_
_
_
_
c
−1
c
0
c
1
_
_
_
_
=
_
_
_
_
0
1
0
_
_
_
_
Hence,
_
_
_
_
c
−1
c
0
c
1
_
_
_
_
=
_
_
_
_
−0.3409
1.1364
−0.2273
_
_
_
_
The output of the equalizer is :
q
m
=
_
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
_
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
¸
_
0 m ≤ −3
c
−1
x
−1
= −0.1023 m = −2
0 m = −1
1 m = 0
0 m = 1
c
1
x
1
= −0.0455 m = 2
0 m ≥ 3
Hence, the residual ISI sequence is :
residual ISI = {. . . , 0, −0.1023, 0, 0, 0, −0.0455, 0, . . .}
Problem 9.45
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
42
Problem 9.46
A discrete time transversal ﬁlter equivalent to the cascade of the trasmitting ﬁlter g
T
(t), the channel
c(t), the matched ﬁlter at the receicer g
R
(t) and the sampler, has tap gain coeﬃcients {x
m
}, where
x
m
=
_
¸
¸
_
¸
¸
_
0.9 m = 0
0.3 m = ±1
0 otherwise
The noise ν
k
, at the output of the sampler, is a zero-mean Gaussian sequence with autocorrelation
function :
E[ν
k
ν
l
] = σ
2
x
k−l
, |k −l| ≤ 1
If the Z-transform of the sequence {x
m
}, X(z), assumes the factorization :
X(z) = F(z)F

(z
−1
)
then the ﬁlter 1/F

(z
−1
) can follow the sampler to white the noise sequence ν
k
. In this case the
output of the whitening ﬁlter, and input to the MSE equalizer, is the sequence :
u
n
=

k
I
k
f
n−k
+ n
k
where n
k
is zero mean Gaussian with variance σ
2
. The optimum coeﬃcients of the MSE equalizer,
c
k
, satisfy :
1

n=−1
c
n
Γ
kn
= ξ
k
, k = 0, ±1
where :
Γ(n −k) =
_
_
_
x
n−k
+ σ
2
δ
n,k
, |n −k| ≤ 1
0 otherwise
ξ
(
k) =
_
_
_
f
−k
, −1 ≤ k ≤ 0
0 otherwise
With
X(z) = 0.3z + 0.9 + 0.3z
−1
= (f
0
+ f
1
z
−1
)(f
0
+ f
1
z)
we obtain the parameters f
0
and f
1
as :
f
0
=
_
_
_
±

0.7854
±

0.1146
, f
1
=
_
_
_
±

0.1146
±

0.7854
The parameters f
0
and f
1
should have the same sign since f
0
f
1
= 0.3. However, the sign itself does
not play any role if the data are diﬀerentially encoded. To have a stable inverse system 1/F

(z
−1
),
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
43
we select f
0
and f
1
in such a way that the zero of the system F

(z
−1
) = f
0
+f
1
z is inside the unit
circle. Thus, we choose f
0
=

0.1146 and f
1
=

0.7854 and therefore, the desired system for the
equalizer’s coeﬃcients is
_
_
_
_
0.9 + 0.1 0.3 0.0
0.3 0.9 + 0.1 0.3
0.0 0.3 0.9 + 0.1
_
_
_
_
_
_
_
_
c
−1
c
0
c
1
_
_
_
_
=
_
_
_
_

0.7854

0.1146
0
_
_
_
_
Solving this system, we obtain
c
−1
= 0.8596, c
0
= 0.0886, c
1
= −0.0266
Problem 9.47
(a) The spectrum of the band limited equalized pulse is
X(f) =
_
_
_
1
2W

n=−∞
x(
n
2W
)e
−j
πnf
W
|f| ≤ W
0 otherwise
=
_
_
_
1
2W
_
2 + 2 cos
πf
W
_
|f| ≤ W
0 otherwise
=
_
_
_
1
W
_
1 + 1 cos
πf
W
_
|f| ≤ W
0 otherwise
where W =
1
2T
b
(b) The following table lists the possible transmitted sequences of length 3 and the corresponding
output of the detector.
-1 -1 -1 -4
-1 -1 1 -2
-1 1 -1 0
-1 1 1 2
1 -1 -1 -2
1 -1 1 0
1 1 -1 2
1 1 1 4
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
45
But
P(|y
m
| > 1|b
m
= 0) =
2

2πσ
ν

1
e
−x
2
/2σ
2
ν
dx
Therefore,
P(e) <
14
8
Q

1
σ
ν

Problem 9.48
Since the partial response signal has memory length equal to 2, the corresponding trellis has 4
states which we label as (I
n−1
, I
n
). The following ﬁgure shows three frames of the trellis. The
labels of the branches indicate the output of the partial response system. As it is observed the free
distance between merging paths is 3, whereas the Euclidean distance is equal to
d
E
= 2
2
+ 4
2
+ 2
2
= 24
u
u
u
u u
u
u
u u
u
u
u u
u
u
u

E E
r
r
r
r
rj
¨
¨
¨
¨
¨B
E

¨
¨
¨
¨
¨B
d
d
d
d

r
r
r
r
rj
r
r
r
r
rj
E
-2
0
-2
-4 -4 -4
(1,1)
(1,-1)
(-1,1)
(-1,-1)
(I
n−1
, I
n
)
Problem 9.49
(a) X(z) = F(z)F

(z
−1
) =
1
2
z + 1 +
1
2
z
−1
. Then, the covariance matrix Γ is :
Γ =

1 + N
0
1/2 0
1/2 1 + N
0
1/2
0 1/2 1 + N
0

and ξ =

1/

2
1/

2
0

PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
46
The optimum equalizer coeﬃcients are given by :
C
opt
= Γ
−1
ξ
=
1
det(Γ)

(1 + N
0
)
2
−1/4 −
1
2
(1 + N
0
) 1/4

1
2
(1 + N
0
) (1 + N
0
)
2

1
2
(1 + N
0
)
1/4 −
1
2
(1 + N
0
) (1 + N
0
)
2
−1/4

1/

2
1/

2
0

=
1

2 det(Γ)

N
2
0
+
3
2
N
0
+
1
4
N
2
0
+
3
2
N
0
+
1
2

N
0
2

1
4

where det(Γ) = (1 + N
0
)

(1 + N
0
)
2

1
2

(b)
det(Γ −λI) = (1 + N
0
−λ)

(1 + N
0
−λ)
2

1
2

λ
1
= 1 + N
0
, λ
2
=
1

2
+ 1 + N
0
, λ
3
= 1 −
1

2
+ N
0
and the corresponding eigenvectors are :
v
1
=

−1/

2
0
1/

2

, v
2
=

1/2
1/

2
1/2

, v
3
=

1/2
−1/

2
1/2

(c)
J
min
(K)|
K=1
= J
min
(1) = 1 −ξ

Γ
−1
ξ =
2N
3
0
+ 4N
2
0
+ 2N
0
+ 3/4
2N
3
0
+ 4N
2
0
+ 5N
0
+ 1
(d)
γ =
1 −J
min
(1)
J
min
(1)
=
2N
2
0
+ 3N
0
+ 3/4
2N
3
0
+ 4N
2
0
+ 1/4
Note that as N
0
→0, γ →3. For N
0
= 0.1, γ = 2.18 for the 3-tap equalizer and γ =

1 +
2
N
0
−1 =
3.58, for the inﬁnite-tap equalizer (as in example 10-2-1). Also, note that γ =
1
N
0
= 10 for the case
of no intersymbol interference.
Problem 9.50
For the DFE we have that :
ˆ
I
k
=
0

j=−K
1
c
j
u
k−j
+
K
2

j=1
c
j
I
k−j
, and ǫ
k
= I
k

ˆ
I
k
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
48
where ψ
lj
=

−l
m=0
f

m
f
m+l−j
+ N
0
δ
lj
, which is the desired expression for the feedforward taps of
the equalizer.
Problem 9.51
The tap coeﬃcients for the feedback section of the DFE are given by the equation :
c
k
= −

0
j=−K
1
c
j
f
k−j
, k = 1, 2, ..., K
2
= −(c
0
f
k
+ c
−1
f
k+1
+ ... + c
−K
1
f
k+K
1
)
But f
k
= 0 for k < 0 and k > L. Therefore :
c
L
= −c
0
f
L
, c
L+1
= 0, c
L+2
= 0, etc
Problem 9.52
(a) The tap coeﬃcients for the feedback section of the DFE are given by the equation : c
k
=

0
j=−K
1
c
j
f
k−j
, 1 ≤ k ≤ K
2
, and for the feedforward section as the solution to the equations :

0
j=−K
1
c
j
ψ
lj
= −f

−l
, K
1
≤ l ≤ 0. In this case, K
1
= 1, and hence :

0
j=−K
1
c
j
ψ
lj
= −f

−l
, l =
−1, 0 or :
ψ
0,0
c
0
+ ψ
0,−1
c
−1
= f

0
ψ
−1,0
c
0
+ ψ
−1,−1
c
−1
= f

1
But ψ
lj
=

−l
m=0
f

m
f
m+l−j
+ N
0
δ
lj
, so the above system can be written :

1
2
+ N
0
1
2
1
2
1 + N
0

c
0
c
−1

=

1/

2
1/

2

so :

c
0
c
−1

=
1

2

N
2
0
+
3
2
N
0
+
1
4

1
2
+ N
0
N
0

2
2

2N
0

, for N
0
<< 1
The coeﬃcient for the feedback section is :
c
1
= −c
0
f
1
= −
1

2
c
0
≈ −1, for N
0
<< 1
(b)
J
min
(1) = 1 −
0

j=−K
1
c
j
f
−j
=
2N
2
0
+ N
0
2

N
2
0
+
3
2
N
0
+
1
4
≈ 2N
0
, for N
0
<< 1
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.
49
(c)
γ =
1 −J
min
(1)
J
min
(1)
=
1 + 4N
0
2N
0
(1 + 2N
0
)

1
2N
0
, N
0
<< 1
(d) For the inﬁnite tap DFE, we have from example 9-5-1 :
J
min
=
2N
0
1+N
0
+

(1+N
0
)
2
−1
≈ 2N
0
, N
0
<< 1
γ

=
1−J
min
J
min
=
1−N
0

(1+N
0
)
2
−1
2N
0
(e) For N
0
= 0.1, we have :
J
min
(1) = 0.146, γ = 5.83 (7.66 dB)
J
min
= 0.128, γ

= 6.8 (8.32 dB)
For N
0
= 0.01, we have :
J
min
(1) = 0.0193, γ = 51 (17.1 dB)
J
min
= 0.0174, γ

= 56.6 (17.5 dB)
The three-tap equalizer performs very wee compared to the inﬁnite-tap equalizer. The diﬀerence
in performance is 0.6 dB for N
0
= 0.1 and 0.4 dB for N
0
= 0.01.
Problem 9.53
(a) We have that :
1
2T
= 900,
1+β
2T
= 1200 ⇒
1 + β = 1200/900 = 4/3 ⇒β = 1/3
(b) Since 1/2T = 900, the pulse rate 1/T is 1800 pulses/sec.
(c) The largest interference is caused by the sequence : {1, −1, s, 1, −1, 1} or its opposite in sign.
This interference is constructive or destructive depending on the sign of the information symbol s.
The peak distortion is

3
k=−2,k=0
f
k
= 1.6
(d) The probability of the worst-case interference given above is

1
2

5
= 1/32, and the same is the
PROPRIETARY MATERIAL. c The McGraw-Hill Companies, Inc. All rights reserved. No part of this Manual may be displayed,
reproduced or distributed in any form or by any means, without the prior written permission of the publisher, or used beyond the
limited distribution to teachers and educators permitted by McGraw-Hill for their individual course preparation. If you are a
student using this Manual, you are using it without permission.

|f | < 1200. |f | < 1200 Thus. the frequency response of the transmitting ﬁlter is GT (f ) = √ T . the the signal pulse used for spectral shaping. should have roll-oﬀ factor β = 0 with corresponding spectrum : X(f ) = T. then the symbol rate is R= 4800 = 2400 symbols/sec 2 In order to satisfy the Nyquist criterion. .15 (b) If the bit rate is 4800 bps.

34 (a) Taking the inverse Fourier transform of H(f ). y(t) = u(t) ⋆ h(t) = ∞ ∞ n=−∞ In s(t − nT ) ∞ ⋆ δ(t) + α α δ(t − t0 ) + δ(t + t0 ) 2 2 ∞ α α In s(t − t0 − nT ) + In s(t + t0 − nT ) In s(t − nT ) + = 2 n=−∞ 2 n=−∞ n=−∞ . we obtain : h(t) = F −1 [H(f )] = δ(t) + Hence. Hence.Problem 9. y(t) = s(t) ⋆ h(t) = s(t) + α α δ(t − t0 ) + δ(t + t0 ) 2 2 α α s(t − t0 ) + s(t + t0 ) 2 2 (b) If the signal s(t) is used to modulate the sequence {In }. then the transmitted signal is : u(t) = ∞ n=−∞ In s(t − nT ) The received signal is the convolution of u(t) with h(t).

we obtain : wk = Ik x0 + n=k In xk−n In xk−n−1 + n=k α α + Ik x−1 + 2 2 = Ik x0 + α α Ik x1 + 2 2 In xk−n+1 n=k α α x−1 + x1 + 2 2 In xk−n + n=k α α xk−n−1 + xk−n+1 2 2 The terms under the summation is the ISI introduced by the channel. the output of the matched ﬁlter s(−t) at the time instant t1 is : w(t1 ) = ∞ n=−∞ ∞ In ∞ −∞ s(τ − nT )s(τ − t1 )dτ ∞ −∞ ∞ −∞ α + In 2 n=−∞ α In + 2 n=−∞ ∞ s(τ − t0 − nT )s(τ − t1 )dτ s(τ + t0 − nT )s(τ − t1 )dτ If we denote the signal s(t) ⋆ s(t) by x(t). then : xk = 0. k = 0 and the aobove expression simpliﬁes to : wk = Ik + α (Ik+1 + Ik−1 ) 2 . then the output of the matched ﬁlter at t1 = kT is : w(kT ) = + ∞ n=−∞ In x(kT − nT ) ∞ ∞ α α In x(kT − t0 − nT ) + In x(kT + t0 − nT ) 2 n=−∞ 2 n=−∞ (c) With t0 = T and k = n in the previous equation.30 Thus. If the signal s(t) is designed so as to satisfy the Nyquist criterion.

3 0.4762 .9δ(t) + 0.  c−1   c0  c1       = 1     0   0  can be found by solving the previous matrix equation.9 0.41 (a) The equivalent discrete-time impulse response of the channel is : 1 h(t) = n=−1 hn δ(t − nT ) = 0. −0.4286   −0. then the equalized signal is : 1 qm = n=−1 cn hm−n which in matrix notation is written as :   c−1 0.    0.3δ(t − T ) If by {cn } we denote the coeﬃcients of the FIR equalizer.9 0.3δ(t + T ) + 0.Problem 9.3 0.3 0.4762   =    1.9 The coeﬃcients of the zero-force equalizer Thus.3   c0   c1 0. 0.

1429 cn h3−n = 0 q−2 = q3 = n=−1 1 n=−1 1 q−3 = n=−1 cn h−3−n = 0 . ±3 are given by 1 q2 = n=−1 1 cn h2−n = c1 h1 = −0.36 (b) The values of qm for m = ±2.1429 cn h−2−n = c−1 h−1 = −0.

1023 m = −2      0  m = −1   1     0      c x = −0.3   c0   0. 0. .2 1. the residual ISI sequence is : residual ISI = {.1364     −0.0455.0 0. .3409 qm =   0 m ≤ −3      c−1 x−1 = −0.1023.Problem 9.0 0. −0.2273 −0.0 c1 Hence. . −0. 0. .0  c−1     = 1     0   0  The output of the equalizer is :     =  1. . 0.0 0.44 The optimum tap coeﬃcients of the zero-force equalizer can be found by solving the system:     0. 0. 0.    c0  c1 c−1   1. .3 0.0455  1 1     0 m=0 m=1 m≥3 m=2 Hence.2 1.} .

X(z). the matched ﬁlter at the receicer gR (t) and the sampler.1146  √  ± 0. is the sequence : un = k Ik fn−k + nk where nk is zero mean Gaussian with variance σ 2 .46 A discrete time transversal ﬁlter equivalent to the cascade of the trasmitting ﬁlter gT (t).3. where   0.1146 f1 = √  ± 0.9 m = 0   xm = 0. ck . In this case the output of the whitening ﬁlter. |k − l| ≤ 1 If the Z-transform of the sequence {xm }. and input to the MSE equalizer. To have a stable inverse system 1/F ∗ (z −1 ). |n − k| ≤ 1 Γ(n − k) =  0 otherwise   f . −1 ≤ k ≤ 0 −k ξ( k) =  0 otherwise X(z) = 0. the channel c(t). ±1 where :   x 2 n−k + σ δn. assumes the factorization : X(z) = F (z)F ∗ (z −1 ) then the ﬁlter 1/F ∗ (z −1 ) can follow the sampler to white the noise sequence νk . f0 = √  ± 0.3z −1 = (f0 + f1 z −1 )(f0 + f1 z) we obtain the parameters f0 and f1 as :  √  ± 0.9 + 0. the sign itself does not play any role if the data are diﬀerentially encoded. n=−1 k = 0.3z + 0. is a zero-mean Gaussian sequence with autocorrelation function : E[νk νl ] = σ 2 xk−l . at the output of the sampler. .3 m = ±1    0 otherwise The noise νk .7854 With The parameters f0 and f1 should have the same sign since f0 f1 = 0. satisfy : 1 cn Γkn = ξk .42 Problem 9. The optimum coeﬃcients of the MSE equalizer. However.k . has tap gain coeﬃcients {xm }.7854 .

7854 c−1 0.9 + 0. c0 = 0. we choose f0 = 0.9 + 0.0 0.9 + 0.0266 . we obtain c−1 = 0.3 0.0        c0  =  √0.1 0. the desired system for the equalizer’s coeﬃcients is      √ 0.1146   0.we select f0 and f1 in such a way that the zero of the system F ∗ (z −1 ) = f0 + f1 z is inside the unit √ √ circle. Thus.1 0 Solving this system.3      c1 0.0886.7854 and therefore. c1 = −0.8596.3 0.1 0.3 0.1146 and f1 = 0.

49 1 1 (a) X(z) = F (z)F ∗ (z −1 ) = 2 z + 1 + 2 z −1 . Then. the covariance matrix Γ is :  Γ=   1 + N0 1/2 0 1/2 1 + N0 1/2 √ 1/ 2   √ 1/2  and ξ =  1/ 2   1 + N0 0 0       .Problem 9.

Also. γ = 2. v2 =  1/ 2 v1 =  0    √ 1/ 2 1/2 (c)  1/2     . λ2 = det(Γ − λI) = (1 + N0 − λ) (1 + N0 − λ)2 − 1 √ 2 1 2 + 1 + N0 .46 The optimum equalizer coeﬃcients are given by : Copt = Γ−1 ξ = 1 det(Γ)     (1 + N0 )2 − 1/4 − 1 (1 + N0 ) 2 1 − 2 (1 + N0 ) = where det(Γ) = (1 + N0 ) (1 + N0 )2 − (b)   √ 1 2 det(Γ)   1/4 2 N0 2 N0 + 3 N0 + 2 + 3 N0 + 2 1 4 1 2 1 4 1 2 √ 1/ 2  √ 1 (1 + N0 )2 − 2 (1 + N0 )   1/ 2  1 2 − 1/4 0 − 2 (1 + N0 ) (1 + N0 )  1/4         − N0 − 2 λ1 = 1 + N0 . γ → 3.58. v3 =  −1/√2     1/2  2 3 2N0 + 4N0 + 2N0 + 3/4 3 + 4N 2 + 5N + 1 2N0 0 0 Jmin (K)|K=1 = Jmin (1) = 1 − ξ ′ Γ−1 ξ = (d) γ= 2 1 − Jmin (1) 2N0 + 3N0 + 3/4 = 3 2 Jmin (1) 2N0 + 4N0 + 1/4 3. λ3 = 1 −  1 √ 2 + N0 ⇒ and the corresponding eigenvectors are :   √  −1/ 2 1/2    √  . For N0 = 0.18 for the 3-tap equalizer and γ = 1 N0 1+ 2 N0 = 10 for the case −1 = . for the inﬁnite-tap equalizer (as in example 10-2-1).1. note that γ = of no intersymbol interference. Note that as N0 → 0.

0 or : ∗ ψ0. for N0 << 1 2 (b) 0 Jmin (1) = 1 − j=−K1 cj f−j = 2 2N0 + N0 2 + 3N + 2 N0 2 0 1 4 ≈ 2N0 . K1 = 1.−1 c−1 = f1 But ψlj = −l ∗ m=0 fm fm+l−j so :   c0 c−1  + N0 δlj .0 c0 + ψ0. l = j=−K1 cj ψlj = −f−l . K1 ≤ l ≤ 0. for N0 << 1 ≈ √ 2 2N0  √  1 c1 = −c0 f1 = − √ c0 ≈ −1.0 c0 + ψ−1.Problem 9. 1 ≤ k ≤ K2 . for N0 << 1 . and for the feedforward section as the solution to the equations : 0 0 ∗ ∗ j=−K1 cj ψlj = −f−l .52 (a) The tap coeﬃcients for the feedback section of the DFE are given by the equation : ck = − 0 j=−K1 cj fk−j . In this case. so the above system can be written :    √   1 1 c0 1/ 2 + N0 2  = √   2 1 1 + N0 c−1 1/ 2 2 1 4 The coeﬃcient for the feedback section is : 1 = √ 2 + 3N + 2 N0 2 0   1 2 + N0 N0  2  . and hence : −1.−1 c−1 = f0 ∗ ψ−1.

N0 << 1 Jmin (1) 2N0 (1 + 2N0 ) 2N0 .(c) γ= 1 − Jmin (1) 1 + 4N0 1 = ≈ .

scribd
/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->