DEPARTMENT OF ELECTRICAL ENGINEERING

NATIONAL INSTITUTE OF TECHNOLOGY, ROURKELA

Assignments-1, Session: 2023 - 2024 (Spring)
B. Tech 6th Semester
Communication Systems (Code: EE3402)

Answer all questions

1. 3.8-3 Find the mean square value (or power) of the output voltage y(t) of the RC network shown
in Fig. P3.5-4 with RC = 2π if the input voltage PSD Sx (f ) is given by (a) K; (b) Π(πf ); (c)
[δ(f + 1) + δ(f − 1)]. In each case calculate the power (mean square value) of the input signal x(t).
R

+ g(t) C y(t)
−

2. 3.8-4 Find the mean square value (or power) of the output voltage y(t) of the system shown in Fig. P3.84
if the input voltage PSD Sx (f ) = Π(πf ). Calculate the power (mean square value) of the input signal
1Ω

+ x(t) d
y(t)
− dt
1F
−

x(t).

3. 3.7-4 Show that (

∞
m ̸= n
Z
0
sinc(2πBt − mπ) sinc(2πBt − nπ)dt = 1
−∞ 2B m=n

Hint. Recognize that

    
k 1 f
sinc(2πBt − kπ) = sinc 2πB t − ⇐⇒ Π, e−jπf k/B
2B 2B 2B

Use this fact and the result in Prob. 3.7-2 to show that
Z ∞ Z B
1
sinc(2πBt − mπ) sinc(2πBt − nπ)dt = ej(n−m)/2B)2πf df
−∞ 4B 2 −B

The desired result follows from this integral.

4. 3.7-5 For the signal
2a
g(t) =
t2 + a2
determine the essential bandwidth B Hz of g(t) such that the energy contained in the spectral components
of g(t) of frequencies below B Hz is 99% of the signal energy Eg .
Hint. Determine G(f ) by applying the duality property [Eq. (3.26)] to pair 3 of Table 3.1.

5. 3.7-6 A low-pass signal g(t) is applied to a squaring device. The squarer output g 2 (t) is applied to a unity
gain ideal low-pass filter of bandwidth ∆f Hz (Fig. P3.7-6). Show that if ∆f is very small (∆f → 0),
the filter output is a dc signal of amplitude 2Eg ∆f , where Eg is the energy of g(t). Hint: The output
y(t) is a de signal because its spectrum Y (f ) is concentrated at f = 0 from −∆f to ∆f with ∆f → 0
(impulse at the origin). If g 2 (t) ⇐⇒ A(f ), and y(t) ⇐⇒ Y (f ), then Y (f ) ≈ [2A(0)∆f ]δ(f ). Now, show
that Eg = A(0).
g(t) g 2 (t) y(t) = 2Eg ∆f
()2 LPF

g(t) = C cos (2πf0 t + θ0 ) is given by Rg (τ ) = C 2 /2 cos 2πf0 τ ,

6. 3.8-1 Show that the autocor relation function of
and the corresponding PSD is Sg (f ) = C 2 /4 [δ (f − f0 ) + δ (f + f0 )]. Hence, show that for a signal y(t)

given by
X∞
y(t) = C0 + Cn cos (n2πf0 t + θn )
n=1
the autocorrelation function and the PSD are given by
∞
1X 2
Ry (τ ) = C02 + Cn cos n2πf0 τ
2
n=1
∞
1X
Sy (f ) = C02 δ(f ) + Cn 2 [δ (f − nf0 ) + δ (f + nf0 )]
4
n=1

Hint: Show that if g(t) = g1 (t) + g2 (t), then Rg (τ ) = Rg1 (τ ) + Rg2 (τ ) + Rg1 g2 (τ ) + Rg2 g1 (τ ), where
R T /2
Rg1 g2 (τ ) = limT →∞ (1/T ) −T /2 g1 (t)g2 (t + τ )dt. If g1 (t) and g2 (t) represent any two of the infinite terms
in y(t), then show that Rg1 g2 (τ ) = Rg2 g1 (τ ) = 0. To show this, use the fact that the area under any
sinusoid over a very large time interval is at most equal to the area of the half-cycle of the sinusoid.

7. 3.8-2 The random binary signal x(t) shown in Fig. P3.8-2 transmits one digit every Tb seconds. A binary
1 is transmitted by a pulse p(t) of width Tb /2 and amplitude A; a binary 0 is transmitted by no pulse.
The digits 1 and 0 are equally likely and occur randomly. Determine the autocorrelation function Rx (τ )
and the PSD Sx (f ).
x(t)
1 0 1 0 1

Tb Tb Tb t
2
Tb 2
Tb 2
Tb

8. 3.7-1 Show that the energy of the Gaussian pulse

1 t2
g(t) = √ e− 2σ2
σ 2π
 √
from direct integration is 1/2σ π. Verify this result by using Parseval’s theorem to derive the energy Eg
from G(f ). Hint: See pair 22 in Table 3.1. Use the fact that
Z ∞Z ∞ Z ∞
−x2 −y 2 2 √
e dxdy = π ⇒ e−x dx = π
−∞ −∞ −∞

9. 3.7-2 Show that Z ∞

π
sinc2 (kt)dt =
−∞ 4

10. 3.6-2 The distortion caused by multipath transmission can be partly corrected by a tapped delay-line
equalizer. Show that if α ≪ 1, the distortion in the multipath system in Fig. 3.31a can be approximately
corrected if the received signal in Fig. 3.31a is passed through the tapped delayline equalizer shown in
Fig. P3.6-2.
Hint: From Eq.
(3.64a), it is clear that the equalizer filter transfer function should be Heq (f ) =
1/ 1 + αe−j2πf ∆t . Use the fact that 1/(1 − x) = 1 + x + x2 + x3 + · · · if x ≪ 1 to show what
should be the tap parameters ai to make the resulting transfer function

H(f )Heq (f ) ≈ e−j2πf td

11. 3.5-5 A bandpass signal g(t) of bandwidth B = 2000 Hz centered at f = 105 Hz is passed through
the RC filter in Fig. P3.5-4 with RC = 10−3 . If over the passband, a variation of less than 2% in
amplitude response and less than 1% in time delay is considered distortionless transmission, would g(t)
be transmitted without distortion? Find the approximate expression for the output signal.
R

+ g(t) C y(t)
−

12. 3.6-1 A certain channel has ideal amplitude, but nonideal phase response (Fig. P3.6-1), given by

|H(f )| = 1
θh (f ) = −2πf t0 − k sin 2πf T k≪1

(a) Show that y(t), the channel response to an input pulse g(t) band-limited to B Hz, is
k
y(t) = g (t − t0 ) + [g (t − t0 − T ) − g (t − t0 + T )]
2

Hint: Use e−jk sin 2πf T ≈ 1 − jk sin 2πf T .

13. 3.5-2 Consider a filter with the transfer function

2 −j2πf t
H(f ) = e−k(2πkf ) 0

Show that this filter is physically unrealizable by using the time domain criterion [noncausal h(t)] and the
frequency domain (Paley-Wiener) criterion. Can this filter be made approximately realizable by choosing
a sufficiently large t0 ? Use your own (reasonable)criterion of approximate realizability to determine t0 .
Hint: Use pair 22 in Table 3.1.
14. 3.5-3 Show that a filter with transfer function

2 105


H(f ) = e−j2πf t0
(2πf )2 + 1010

is unrealizable. Can this filter be made approximately realizable by choosing a sufficiently large t0 ?
Use your own (reasonable) criterion of approximate realizability to determine t0 . Hint: Show that the
impulse response is noncausal.

15. 3.5-4 Determine the maximum bandwidth of a signal that can be transmitted through the low-pass RC
filter in Fig. P3.5-4 with R = 1000 and C = 10−9 if, over this bandwidth, the amplitude response (gain)
variation is to be within 5% and the time delay variation is to be within 2%.

16. 3.3-8 A signal g(t) is band-limited to B Hz. Show that the signal g n (t) is band-limited to nB Hz. Hint:
g 2 (t) ⇐⇒ [G(f ) ∗ G(f )], and so on. Use the width property of convolution.

17. Find the Fourier transform of the signal in Fig. P3.3-3 a by three different methods: (a) By direct
integration using the definition (3.9a). (b) Using only pair 17 Table 3.1 and the time-shifting property.
(c) Using the time differentiation and time-shifting properties, along with the fact that δ(t) ⇐⇒ 1.

g(t)
1

t
−4 −2 2 4

−1
t
Hint: 1 − cos 2x = 2 sin2 x.

18. The process of recovering a signal g(t) from the modulated signal g(t) cos 2πf0 t is called demodulation.
Show that the signal g(t) cos 2πf0 t can be demodulated by multiplying it by 2 cos 2πf0 t and passing the
product through a low-pass filter of bandwidth B Hz [the bandwidth of g(t)]. Assume B < f0 .
Hint: 2 cos2 2πf0 t = 1 + cos 4πf0 t. Recognize that the spectrum of g(t) cos 4πf0 t is centered at 2f0 and
will be suppressed by a low-pass filter of bandwidth B Hz.

19. Signals g1 (t) = 104 Π 104 t and g2 (t) = δ(t) are applied at the inputs of the ideal low-pass filters

H1 (f ) = Π(f /20, 000) and H2 (f ) = Π(f /10, 000) (Fig.). The outputs y1 (t) and y2 (t) of these filters are
multiplied to obtain the signal y(t) = y1 (t)y2 (t). (a) Sketch G1 (f ) and G2 (f ). (b) Sketch H1 (f ) and
H2 (f ). (c) Sketch Y1 (f ) and Y2 (f ). (d) Find the bandwidths of y1 (t), y2 (t), and y(t).

g1 (t) y1 (t)
H1 (f )
y(t) = y1 (t) · y2 (t)
×
g2 (t) y2 (t)
H2 (f )

20. Use the frequency shift property and Table to find the inverse Fourier transform of the spectra shown in
Fig. below. Notice that this time, the Fourier transform is in the ω domain.
 G(ω)

ω
−10 −5 −3 3 5 10

G(ω)

ω
−10 −6 −2 2 6 10

g(t) G(f )
1 e−at u(t) 1
a+j2πf a>0
1
2 eat u(−t) a−j2πf a>0
3 e−a|t| 2a
a2 +(2πf )2
a>0
4 te−at u(t) 1
(a+j2πf )2
a>0
5 tn e−at u(t) n!
(a+j2πf )n+1
a>0
6 δ(t) 1
7 1 δ(f )
8 ej2πf0 t δ (f − f0 )
9 cos 2πf0 t 0.5 [δ (f + f0 ) + δ (f − f0 )]
10 sin 2πf0 t j0.5 [δ (f + f0 ) − δ (f − f0 )]
1 1
11 u(t) 2 δ(f ) + j2πf
2
12 sgn t j2πf
1 j2πf
13 cos 2πf0 tu(t) 4 [δ (f − f0 ) + δ (f + f0 )] + (2πf0 )2 −(2πf )2
1 2πf0
14 sin 2πf0 tu(t) 4j [δ (f − f0 ) − δ (f + f0 )] + (2πf0 )2 −(2πf )2
17 Π τt


τ sinc(πf
  τ)
f
18 2B sinc(2πBt) Π 2B  
2 πf τ
19 ∆ τt τ


sinc
2   2
f
20 B sinc2 (πBt) ∆ 2B
P∞
f0 ∞ 1
P
21 n=−∞ δ(t − nT ) √ n=−∞
δ (f − nf0 ) f0 = T
−t2 2 −2(σπf )2
22 e /2σ σ 2πe
 g(t) G(ω)
1 e−at u(t) 1
a+jω a>0
1
2 eat u(−t) a−jω a>0
3 e−a|t| 2a
a2 +(ω)2
a>0
4 te−at u(t) 1
(a+jω)2
a>0
5 tn e−at u(t) n!
(a+jω)n+1
a>0
6 δ(t) 1
7 1 δ(f )
8 ejω0 t δ (ω − ω0 )
9 cos ω0 t 0.5 [δ (ω + ω0 ) + δ (ω − ω0 )]
10 sin ω0 t j0.5 [δ (ω + ω0 ) − δ (ω − ω0 )]
1 1
11 u(t) 2 δ(f ) + jω
2
12 sgn t jω
1 jω
13 cos ω0 tu(t) 4 [δ (ω − ω0 ) + δ (ω + ω0 )] + (ω0 )2 −(ω)2
1 ω0
14 sin ω0 tu(t) 4j [δ (ω − ω0 ) − δ (ω + ω0 )] + (ω0 )2 −(ω)2
17 Π τt


τ sinc(ωτ
/2)
ω
18 2W sinc(W t) 2πΠ 2W
19 ∆ τt τ 2 ωτ



2 sinc 4

2 ω
20 2W
P∞sinc (W t) 2π∆P∞4W 1
21 n=−∞ δ(t − nT ) f0 n=−∞ δ (f − nf0 )
√
f0 = T
−t 2 2 2
22 e /2σ σ 2πe−2(σπf )