Solutions

Download as pdf or txt
Download as pdf or txt
You are on page 1of 9

Ziemer & Tranter: Principles of Communications

Problem Solutions
Chapter 2
Problem 2.2:

x(t) = 8 sin(8t) + 4 cos(4t /4).

Problem 2.4:

(a)

Single-sided spectrum
|Xa(f)|

Double-sided spectrum
|Xa(f)|

f, Hz
0
0

3
3

f, Hz

f, Hz
3 0
Xa(f)

Xa(f)

/3

(b)

/3
0 3
3
/3

Single-sided spectrum

|Xb(f)|

4
2

Z
Problem 2.6:

Area = 2
0

2
1

f, Hz
5
5

6
6

f, Hz

Xb(f)

/2

f, Hz

Double-sided spectrum

|Xb(f)|

0
0

f, Hz

65
0
Xb(f) /2
0
65
/2

56
56
f, Hz

2
1
(1 t/)dt = ( /2) = 1.



(b) 380;

(d) 12 e5/2 ;

(c) 0;

(e) 4 2 + 100.

Problem 2.8:

(a) 65;

Problem 2.10:

(a) x(t) = 2Re[ej5t ] + 4Re[ej(11t/2) ] = Re[2ej5t + 4ej(11t/2) ].


(b) x(t) = ej5t + ej5t + 2ej(11t/2) + 2ej(11t/2) .
(c)

Single-sided spectrum
4

|Xb(f)|

|X(f)|

2
f, Hz
0
0

/2

Problem 2.12:

Problem 2.14:

(a) Power (P =

Double-sided spectrum

2.5
2.5

Xb(f)
A2
);
2

5.5
5.5

f, Hz

2
1
5.5 2.5 0

2.5
/2
0
2.5

f, Hz
5.5
5.5

5.5 2.5
/2 Xb(f)

f, Hz

(b) Neither (E = and P = 0);

(c) Energy (E = 34 A2 5 );

(d) Energy (E = 5 ).

(a) E = , P = ;

(b) E = 5 J, P = 0;

(c) E = , P = 32 W;

(d) E = , P = 8.5 W.

Problem 2.16:

(a) Break the integral up into contiguous intervals of length T0 .

Problem 2.18:

(a) Expand the integrand, integrate term by term, and simplify, making use
of the orthogonality property of the orthonormal functions.
(b) Add and subtract the quantity suggested right above (2.31) and simplify.

Problem 2.20:

(a) X2 = 14 , X0 = 21 , X2 = 14 , all other coefficients zero.


(b) X1 = 21 (1 + j), X1 = 12 (1 j), all other coefficients zero.
(c) X4 = X4 = 18 , X2 = X2 = 14 , X0 = 14 , all other coefficients zero.
(d) X3 = 18 , X1 = 38 , X1 = 38 , X3 = 18 , all other coefficients zero.

Problem 2.22:

Parts (a) through (c) were discussed in the text.


For (d), break the integral for x(t) up into a part for t < 0 and a part for t > 0.
Then apply the odd half-wave symmetry condition.

Problem 2.24:

(a) P|nf0 |1/ /Ptotal = 0.905;

(b) P|nf0 |1/ /Ptotal = 0.903.

Problem 2.26:

(a) 2 /8;

(b) /4 (Use Fourier series of pulse train).

Problem 2.28:

(a) There is no line at DC; otherwise, it looks like a square wave spectrum.

Problem 2.30:
Problem 2.32:
Problem 2.34:

(b) XnA = K(jn0 )XnB , where the superscripts A and B refer to xa and xb .
1
1
+ (f ).
By linearity, F[u(t)] =
j2f
2
The proof is similar to the proof of the convolution theorem.
2

4/3
;
(b) G2 (f ) = 49 (f /30);
(a) G1 (f ) =
2
1 + [f /(3/2)]
2
4/9
1.5

G (f)
2

G (f)
1

1
0.5
0
5

f
5

(c) G3 (f ) =

16
sinc2 (f /5);
25

f
15

 
(d) G4 (f ) = 0.4 sinc

0
f 20
5

15

+ sinc

f +20
5

 2

0.8
G4(f)

0.16

G (f)

0.6

0.4
0.2
0
15 10 5

Problem 2.36:

(a) y1 (t) =

0,

10

f
15

f
20

20

t < 0.5



1 e(t +0.5) ,
0.5 < t < + 0.5


1
e(t 0.5) e(t +0.5) , t > + 0.5

(b) Plot of y2 (t):

(c) Normalised plot of y3 (t) for various values of :

2
2/
=50

y2(t)

1.5

=10
y3(t)

1
=2

=1

0.5
0

1.5 1 0.5 0 0.5 1 1.5

0.5

0.5

E1 (|f |W )
Etotal

=2

(d) y4 (t) =

x()d.

E1 (|f |W )
Etotal

tan1 (2W/);

R W

sinc2 (u)du.

Problem 2.38:

(a)

Problem 2.40:

Combine the exponents of the two factors in the integrand of the Fourier

(b)

transform integral, complete the square, and use the definite integral given.
Problem 2.42:

The result is an even triangular wave with zero average value of period T0 .
It makes no difference whether the original square wave is even or odd or neither.

Problem 2.44:

(a) h(t) = (t) 10e10t u(t);

(b) h(t) = 0.3e0.4(t3) u(t 3).

Problem 2.46:

(a) Replace the capacitors with 1/(jC) which is their ac-equivalent impedance.
Call the junction of the input resistor, feedback resistor and capacitors 1. Call
the junction at the positive input of the op amp 2. Call the junction at the
negative input of the op amp 3. Write down the KCL equations at these three
junctions. Use the constraint equation for the op amp V2 = V3 and the
definitions for 0 , Q, and K to get the transfer function expression.
(d) Combinations of components giving RC = 2.3 104 s and

Ra
Rb

= 2.5757 will work.

f 2
The Paley-Wiener criterion gives the non-convergent integral I =
df .
2
1 + f
Z

Problem 2.48:

1
;
4+(2f )2

Problem 2.50:

Input ESD Gx (f ) =

Problem 2.52:

(a) B90 =

Problem 2.54:

(a) Amplitude distortion only;

tan(0.45) = 1.005;

(c) Distortionless;
Problem 2.56:

Output ESD Gy (f ) =
(b) B90 = 0.9W ;

(c) B90 = 0.85/ .

(b) Phase distortion only;


(d) Distortionless.
0.333
.
1+(0.667f )2

The group delay is Tg (f ) =

0.1
1+(0.2f )2

The phase delay is Tp (f ) =

1
[tan1 (0.2f )
2f

(1.075)2 |H(1000)|2
(0.025)2 |H(3000)|2

400
.
[9+(2f )2 ]2 [4+(2f )2 ]

tan1 (0.667f )].


[4 + 16 106 Q2 ][1 + 144 106 Q2 ]
= 1849
.
[4 + 144 106 Q2 ][(1 + 32 106 Q2 )2 + 16 106 Q2 ]

Problem 2.58:

(a) THD =

Problem 2.60:

(b) By trial and error, Q = 0.55 gives 0.01% THD.


1
, where R = f /f3 , and R if f > f3 and R 0 if f < f3 .
(a) |HBU (f )| =
1 + R2n

(b) The ripple in dB is given by RdB = 20 log10 1 + 2 .

Problem 2.62:

|HCH (n )|2 =

1
= HCH (sn )HCH (sn )|sn =jn , where n = /3dB .
1 + 2 (2n2 1)2

Factor the expression and take the left half plane poles for HCH (f ).
Then let sn = jn . Differentiate the argument of this expression
with respect to frequency to get the group delay.
Problem 2.64:

(a) 0.5 second; Use sketches to show (b) and (c).

Problem 2.66:

(a) The sampling frequency should be large compared with the signal bandwidth.

X
(b) Y (f ) = sinc(Ts f )
X(f nfs )ejTs f , where fs = 1/Ts .
n=

For small distortion we want Ts much less than the inverse signal bandwidth,
which implies that we are sampling much faster than the Nyquist rate.
Problem 2.68:
Problem 2.70:

The lowpass recovery filter can cutoff in the range 1.9+ kHz to 2.1 kHz.
Z
cos 0 (t )
Use the form cos0 t =
d and expand the cosine to carry out the

integral. Make use of the fact that the integral of sinc(x) from to is unity.

Problem 2.72:

This is a matter of finding the integral of the product of the signal and its
Hilbert transform, and showing that it is zero.

Problem 2.74:

The spectra for the various signals are shown below, on normalised axes:
1

1.5

0.5

0
2

0
f/W

0
0

(d)

X4(f)/A

X3(f)/A

0.5

Problem 2.76:

f0 f0+W
f

1
(c)

0
2

(b)

X2(f)/A

X1(f)/A

(a)

0.5

0
f/W

0
2

0
f/W

For t < /2. the output is zero. For |t| /2, the result is

/2
cos[2(f0 + f )t ] e(t+ /2) cos[2(f0 + f )t + ] .
y(t) = p
2 + (2f )2
For t > /2, the result is

(/2)et
e /2 cos[2(f0 + f )t ] e /2 cos[2(f0 + f )t + ] .
y(t) = p
2 + (2f )2


2f
1
In the above equations, is given by = tan
.

Chapter 3
Problem 3.2:

Multiply the AM signal xc (t) = Ac [1 + amn (t)] cos c t by 2 cos[c t + (t)] and lowpass
filter it to remove the double frequency (2c ) terms. This yields a detector output of
yD (t) = Ac [1 + amn (t)] cos (t). For negligible phase error this becomes
yD (t) = Ac + Ac amn (t). If the DC component is removed, the demodulated signal is
proportional to the message signal. This is not done in practice since the whole reason
for using AM is to avoid the necessity for coherent demodulation.

Problem 3.4:

Problem 3.6:

Part

hm2n (t)i

a = 0.7

a=1

(a)

1/3

E = 14.04%

E = 25%

(b)

1/3

E = 14.04%

E = 25%

(c)

E = 32.89%

E = 50%

For the first signal, mn (t) = m(t) and E =

5
6

= 83.33%.

For the second signal, mn (t) = 15 m(t) and E =


Problem 3.8:

(a) mn (t) =

1
6

= 16.67%.

1
16

[7 cos 20t + 9 cos 60] .



 2
1 2 1
(b) hm2n (t)i = 16
[7 + 92 ] = 0.2539.
2
0.25 0.2539
(c) E =
= 0.05969 = 5.969%.
1 + 0.25 0.2539


1
(d) xc (t) = 100 1 + 32
(7 cos 20t + 9 cos 60t) cos 200t
= 14.0625 cos 140t + 10.9375 cos 180t + 100 cos 200t+
10.9375 cos 220t + 14.0625 cos 260t.
All the terms are positive and real so the phase spectrum is zero everywhere.
Given A = 12 (14.0625) = 7.03125 and B = 12 (10.9375) = 5.46875, the amplitude
spectrum is given by
50

50
A=7.031
B=5.469

130 110 100 90

70

70

90

110

130

B = 13.89;

100

Problem 3.10:

A = 30;

a = 0.926.

Problem 3.12:

(a) The minimum of m(t) is M = 3.432, and therefore


mn (t) = 0.5828 cos 2fm t + 0.2914 cos 4fm t + 0.5828 cos 10fm t.
The AM signal is xc (t) = Ac [1 + 0.8mn (t)] cos 2fc t.

The spectrum contains 14 impulses with the following frequencies and amplitudes:
F:

fc 5fm

fc 2fm

fc fm

fc

fc + fm

fc + 2fm

fc + 5fm

A:

0.1166Ac

0.0583Ac

0.1166Ac

0.5Ac

0.1166Ac

0.0583Ac

0.1166Ac

F:

fc 5fm

fc 2fm

fc fm

fc

fc + fm

fc + 2fm

fc + 5fm

A:

0.1166Ac

0.0583Ac

0.1166Ac

0.5Ac

0.1166Ac

0.0583Ac

0.1166Ac

(b) The efficiency is 19.6%.


Problem 3.14:

The upper-sideband filter is HU (f ) = 1 12 sgn(f + fc ).


The SSB modulator output is xU SB (t) = 12 Ac m(t) cos c t 12 Ac m(t)
sin c t.

Problem 3.16:

For the USB SSB case the modulator output is a sinusoid of frequency fc + fm while
for the LSB SSB case the modulator output is a sinusoid of frequency of fc fm .

Problem 3.18:

Let the message signal be m(t) = A cos 1 t + B cos 2 t, so that the VSB waveform is
xc (t) = 12 A cos(c 1 )t + 12 A(1 ) cos(c + 1 )t + 21 B cos(c + 2 )t.
After carrier reinsertion the signal becomes y(t) = xc (t) + K cos c t, or
y(t) = y1 (t) cos c t + y2 (t) sin c t
where y1 (t) =

A
2

cos 1 t +

B
2

cos 2 t + K and y2 (t) = (A + A2 ) sin 1 t

B
2

sin 2 t.

This signal can also be written as y(t) = R(t) cos(c t + ), where R(t) is the envelope
p
given by R(t) = y12 (t) + y22 (t). For large K the output of the envelope detector
becomes R(t) |y1 (t)| = 12 m(t) + K, where the dc bias K can be removed to leave
the output proportional to the message signal m(t).
Problem 3.20:

For high-side tuning the local oscillator frequency is given by fLO = fi + fIF .
fIF + 25
The tuning range of the LO for a given IF frequency is therefore R =
.
fIF + 5
The tuning range R varies from 4.70 to 3.86 as fIF varies from 0.4 MHz to 2.0 MHz.

Problem 3.22:

For high-side tuning


fLO = fi + fIF = 1120 + 2500 = 3620 kHz
fIM AGE = fi + 2fIF = 1120 + 5000 = 6120 kHz
For low-side tuning
fLO = fi fIF = 1120 2500 = 1380 kHz,

Problem 3.24:

fLO = 1380 kHz

fIM AGE = fi 2fIF = 1120 5000 = 3880 kHz, fIM AGE = 3880 kHz

Since P = 21 A2c the carrier amplitude is Ac = 2P = 8.944. Thus the FM signal is


xc (t) = 8.944 cos [2(40)t + 10 sin(2(5)t)]. The signal has a non-zero DC offset since
the frequency deviation f = 50 Hz is larger than the carrier frequency fc = 40 Hz.

This causes the sideband n = 8 at the frequency origin to be significant, with an


amplitude of Ac J8 (10) = 2.843, which also is the average value of xc (t). The nonzero

Amp. spectrum |Xc(f)|

DC value is obvious in the plots of the spectrum |Xc (f )| and signal xc (t) below.
3
2
1

0
150

100

50
0
50
Frequency (Hz)

100

150

Signal xc(t)

10

10
0

Problem 3.26:

0.05

0.1
Time (s)

0.15

0.2

The single-sided amplitude (normalised to Ac ) and phase spectra are shown below.
Amplitude/A

0.4

0.2

Phase (rad)

0
700

900
1000 1100
Frequency (Hz)

1200

1300

800

900
1000 1100
Frequency (Hz)

1200

1300

0
700

The message frequency is fm = 5 Hz and the modulation index is = 5. Since the


filter bandwidth is 54 Hz, it passes the carrier and 5 components to either side.
5
5
X
1 2 X 2
The filter output power is therefore P = Ac
J () = 100
Jn2 (5) = 95.92 W.
2 n=5 n
n=5
The single-sided amplitude and phase spectrum of the filter output is shown below.
Amplitude

6
4
2
0

Phase (rad)

Problem 3.28:

800

fc25

fc
Frequency (Hz)

fc+25

fc25

fc
Frequency (Hz)

fc+25

Problem 3.30:



Write the modulated signal as xc (t) = R(t) cos(c t + (t)) = Re xc (t)ejc t .
It follows that the complex envelope is given by xc (t) = R(t)e(t) . Therefore, since


sin x = cos(x /2) we can write xc (t) = Re ej200t 5ej40t + 10 + 3ej(40t/2) .
Thus xc (t) = (5 cos 40t + 10 + 3 sin 40t) j(5 sin 40t + 3 cos 40t), resulting in
p
R(t) = (10 + 5 cos 40t + 3 sin 40t)2 + (5 sin 40t + 3 cos 40t)2 and
5 sin 40t 3 cos 40t
(t) = tan1
, which can both be further simplified.
10 + 5 cos 40t + 3 sin 40t

Problem 3.32:

(a) The phase deviation (t) is


(t)

1920
0

t
0

(b) The frequency deviation f (t) =

1 d
is
2 dt

240

1/2 d/dt

t
0

(c) The peak frequency deviation is f = 8fd = 240 Hz.


R4
(d) The peak phase deviation is = 2fd 0 (8)dt = 2(30)(8)(4) = 1920 rad.
(e) The output power is P = 12 A2c = 12 (100)2 = 5000 W.
Problem 3.34:

The frequency deviation in Hz is the plot in Fig. 3.81 with the ordinate values
multiplied by 25. The phase deviation is given by
500
400

(t)

250
50
0

Problem 3.36:

The frequency deviation in Hz is the plot in Fig. 3.83 with the ordinate values
multiplied by 10. The phase deviation is given by
40

(t)

20
0
20

Problem 3.38:
Problem 3.40:

(a) 50.2 kHz;


(b) 52 kHz;
(c) 70 kHz;
(d) 250 kHz.
Z
Z
1
1
Jn () =
ej( sin xnx) dx =
[cos( sin x nx) + j sin( sin x nx)] dx.
2
2
The imaginary part of Jn () is zero since the integrand (sin term) is an odd function
Z
1
of x while the limits (, ) are even. Thus Jn () =
cos( sin x nx) dx.
2 Z
1
Since the integrand (cos term) is even we get Jn () =
cos( sin x nx) dx.
0

Problem 3.42:

(a) f = 80 Hz; (b) (t) = 8 sin 20t; (c) = 8; (d) Pin = 50 W, Pout = 16.76 W;
(e) Single-sided spectrum at filter input:

Single-sided spectrum at filter output:

Spectrum at filter input

Spectrum at filter output

Phase (rad)

Amplitude

0
350

400

450
500
550
Frequency (Hz)

600

650

0
350

Problem 3.44:

400

450
500
550
Frequency (Hz)

600

650

0
350

Phase (rad)

Amplitude

400

450
500
550
Frequency (Hz)

600

650

400

450
500
550
Frequency (Hz)

600

650

0
350

FM spectra for constant message frequency fm and varying frequency deviation f :


= 0.5
f
= 1.0
f
= 2.0
f
= 5.0
f
= 10.0
f
fc

Problem 3.46:

The multiplier factor is n =

20
0.05

= 400. The carrier frequency after the multiplier is

fc2 = n(0.11) = 44 MHz. The two permissible local oscillator frequencies are
fLO1 = 100 44 = 56 MHz and fLO2 = 100 + 44 = 144 MHz. The center frequency
of the bandpass filter must be fc = 100 MHz and the bandwidth is
B = 2(20 + 1)10000 = 420 kHz.

You might also like