DEPARTMENT OF ELECTRICAL ENGINEERING

NATIONAL INSTITUTE OF TECHNOLOGY, ROURKELA

Assignments-4, Session: 2023 - 2024 (Spring)
B. Tech 6th Semester
Communication Systems (Code: EE3402)

Answer all questions

1. Sketch ϕF M (t) and ϕP M (t) for the modulating signal m(t) shown in Fig., given ωc = 2π ×107, kf = 104π,
and kp = 25π .

2. A baseband signal m(t) is the periodic sawtooth signal shown in Fig.

(a) Sketch ϕF M (t) and ϕP M (t) for this signal m(t) if ωc = 2π × 106 , kf = 2000π , and kp = π/2.
(b) Show that the PM signal is a signal with constant frequency but periodic phase changes.

Explain why it is necessary to use kp < π in this case for reception purposes. [Note that the PM signal
has a constant frequency but has phase discontinuities corresponding to the discontinuities of m(t).]

3. A periodic message signal m(t) as shown in Figure is transmitted as an angle-modulated signal.

The modulation system has ωc = 2π × 103 rad/s. Let the signal bandwidth of m(t) be approximated by
5 divided by its own period (i.e., its fifth harmonic frequency).

(a) To generate an FM signal with kf = 20π , sketch the frequency-modulated signal ϕF M (t) in the
time domain.
 (b) If a PM signal is generated for m(t) with kp = π/2, sketch the phase-modulated signal ϕP M (t) in
the time domain.

4. Over an interval |t| ≤ 1, an angle-modulated signal is given by

ϕEM (t) = 10cos(13, 000πt + 0.3π)

It is known that the carrier frequency ωc = 12, 000π .

(a) Assuming the modulated signal is a PM signal with kp = 1000, determine m(t) over the interval
|t| ≤ 1.
(b) Assuming the modulated signal is an FM signal with kf = 1000, determine m(t) over the interval
|t| ≤ 1.

5. A periodic message signal m(t) as shown in Figure is to be transmitted by using angle modulation. Its
bandwidth is approximated by 200 Hz. The modulation system has ωc = 4π × 103 rad/s.

(a) If an FM signal with kf = 500π is to be generated, sketch in the FM signal in the time domain.
(b) If a PM signal with kp = 0.25π is to be generated, sketch the PM signal waveform in the time
domain.

6. For a message signal √

m(t) = 3 cos 1000t − 5 cos 1200 2πt

(a) Write expressions (do not sketch) for φPM (t) and φFM (t) when A = 10, ωc = 106 , kf = 1000π, and
kp = 1. For determining φFM (t), use the indefinite integral of m(t); that is, take the value of the
integral at t = −∞ to be 0 .
(b) Estimate the bandwidths of φFM (t) and φPM (t).

7. An angle-modulated signal with carrier frequency ωc = 2π × 106 is

φEM (t) = 10 cos (ωc t + 0.1 sin 2000πt)

(a) Find the power of the modulated signal.

(b) Find the frequency deviation ∆f .
(c) Find the phase deviation ∆ϕ.
(d) Estimate the bandwidth of φEM (t).

8. An angle-modulated signal with carrier frequency ωc = 2π × 106 is

φEM (t) = 5 cos (ωc t + 20 cos 1000πt + 10 sin 4000t)

(a) Find the power of the modulated signal.

(b) Find the frequency deviation ∆f .
 (c) Find the phase deviation ∆ϕ.
(d) Estimate the bandwidth of φEM (t).

9. Given m(t) = sin 2000πt, kf = 5, 000π, and kp = 10,

(a) Estimate the bandwidths of φFM (t) and φPM (t).

(b) Repeat part (a) if the message signal amplitude is doubled.
(c) Repeat part (a) if the message signal frequency is doubled.
(d) Comment on the sensitivity of FM and PM bandwidths to the spectrum of m(t).
2
10. Given m(t) = e−100t , fc = 104 Hz, kf = 500π, and kp = 1.2π.

(a) Find ∆f , the frequency deviation for FM and PM.

(b) Estimate the bandwidths of the FM and PM waves. Hint: Find M (f ) first and find its 3 dB
bandwidth.

11. (a) Show that when m(t) has no jump discontinuities, an FM demodulator followed by an integrator
(Fig. a) forms a PM demodulator. Explain why it is necessary for the FM demodulator to remove
any dc offset before the integrator.
(b) Show that a PM demodulator followed by a differentiator (Fig. b) serves as an FM demodulator
even if m(t) has jump discontinuities or even if the PM demodulator output has dc offset.

12. A periodic square wave m(t) (Fig. a) frequency-modulates a carrier of frequency fc = 10 kHz with
∆f = 1 kHz. The carrier amplitude is A. The resulting FM signal is demodulated, as shown in Fig. b.
Sketch the waveforms at points b, c, d, and e.
13. Let s(t) be an angle-modulated signal that arrives at a receiver,

s(t) = 2 cos[107πt + 2 sin(1000πt + 0.3π) − 3πcos(2000t)]

(a) Find the bandwidth of this FM signal.

(b) If s(t) is sent to an (ideal) envelope detector, find the detector output signal.
(c) If s(t) is first differentiated before the envelope detector, find the detector output signal.
(d) Explain which detector output can be processed to yield the message signal m(t) and find the
message signal m(t) if kf = 200π.

14. A transmitter transmits an AM signal with a carrier frequency of 1470kHz. When a superheterodyne
radio receiver (which has a poor selectivity in its RF-stage bandpass filter) is tuned to 1530kHz, the
signal is heard loud and clear. However, if this radio is cheap, its front-end bandpass filter is not very
selective. Thus, the same signal is also heard (not as strongly) when tuned to another carrier frequency
setting within the AM range of 535-1605 kHz. State, with reasons, at what frequency you will hear this
station. The IF is 455kHz.

15. Consider a superheterodyne FM receiver designed to receive the frequency band of 88 to 108MHz with IF
of 10.7MHz. What is the range of frequencies generated by the local oscillator for this receiver? Analyze
and explain whether it is possible for an FM receiver to receive both a desired FM station and an image
FM station when tuned to the desired frequency.

16. In shortwave AM radio, the IF is also 455kHz. A receiver is designed to receive shortwave broadcasting
of 25 -meter band between 11.6 and 12.1MHz.

(a) Determine the frequency range of the local oscillator for this receiver.
(b) Analyze and explain whether it is possible for this receiver to receive both a desired AM station and
an image station within the same 25 -meter band.

17. (a) Design (the block diagram of) an Armstrong indirect FM modulator to generate an FM carrier
with a carrier frequency of 98.1MHz and ∆f = 75kHz. A narrowband FM generator is available
at a carrier frequency of 100kHz and a frequency deviation ∆f = 10 Hz. The stockroom also has
an oscillator with an adjustable frequency in the range of 10 to 11MHz. There are also plenty of
frequency doublers, triplers, and quintuplers.
(b) Determine the tunable range of the carrier frequency in the design of part (a).

18. Design (the block diagram of) an Armstrong indirect FM modulator to generate an FM carrier with
a carrier frequency of 96MHz and ∆f = 20kHz. A narrowband FM generator with fC = 200kHz and
adjustable ∆f in the range of 9 to 10 Hz is available. The stockroom also has an oscillator with adjustable
frequency in the range of 9 to 10MHz. There are bandpass filters with any center frequency, and only
frequency doublers are available.

19. Design an Armstrong indirect FM modulator in block diagram to generate an FM signal with carrier
96.3 MHz and ∆f = 20.48 kHz. A narrowband FM generator with fc = 150 kHz and ∆f = 10 Hz
is available. Only a limited number of frequency doublers are available as frequency multipliers. In
addition, an oscillator with adjustable frequency from 13 to 14 MHz is also available for mixing, along
with bandpass filters of any specification.

20. Use small-error analysis of PLL to show that a first-order loop [H(s) = 1] cannot track an incoming signal
whose instantaneous frequency varies linearly with time [θi (t) = kt2 ]. This signal can be tracked within
a constant phase if H(s) = (s + a)/s. It can be tracked with zero phase error if H(s) = (s2 + as + b)/s2 .
21. A second-order PLL is implemented with a nonideal loop filter
s+a
H(s) =
s+b

in which b > 0 is very small.

(a) Applying small-signal analysis, determine the transfer function between e(s) and i(s).
(b) Find the steady state PLL phase error for an incoming phase

θi (t) = (ω0 − ωc )t + ϕ0
 DEPARTMENT OF ELECTRICAL ENGINEERING
NATIONAL INSTITUTE OF TECHNOLOGY, ROURKELA
Assignments-5, Session: 2023 - 2024 (Spring)
B. Tech 6th Semester
Communication Systems (Code: EE3402)

Answer all questions

1. Determine the Nyquist sampling rate for the following signals:

(a) 4 sinc(420πt);
(b) 5 sinc2 (6500πt);
(c) sinc(1800πt) + sinc2 (2000πt);
(d) 2 sinc(500πt) sinc(300πt).

2. Figure shows Fourier spectra of signals g1 (t) and g2 (t). Determine the Nyquist sampling rate for signals
g1 (t), g2 (t), g12 (t), g2m (t), and g1 (t)g2 (t).

Hint: Use the frequency convolution and the width property of the convolution.

3. (a) For the G1 (f ) in Figure, find and sketch the spectrum of its ideally and uniformly sampled signals
at the sampling rate of fs = 7500.
(b) For the G2 (f ) in Figure P5.1-2, find and sketch the spectrum of its ideally and uniformly sampled
signals at the sampling rate of fs = 25000.

4. Two signals g1 (t) = 1000Π(4000t) and g2 (t) = 2000∆(8000t) are applied at the inputs of ideal LPFs
H1 (f ) = Π(f /8, 000) and H2 (f ) = Π(f /5, 000) (Fig.). The outputs y1 (t) and y2 (t) of these filters are
multiplied to obtain the signal y(t) = y1 (t)y2 (t). Find the Nyquist rate of y1 (t), y2 (t), and y(t). Use the
convolution property and the width property of convolution to determine the bandwidth of y1 (t)y2 (t).
5. A lowpass signal g(t) sampled at rate of fs > 2B needs reconstruction. The sampling interval is Ts = 1/fs .

(a) If the reconstruction pulse used is  

t 1
p(t) = Π −
Ts 2
specify an equalizer filter E(f ) to recover g(t).
(b) If the reconstruction pulse used is  
t 1
p(t) = ∆ −
Ts 2
specify an equalizer filter E(f ) to recover g(t).
(c) If the reconstruction pulse used is
 
2πt
p(t) = sin [u(t) − u (t − Ts /2)]
Ts

specify an equalizer filter E(f ) to recover g(t).

(d) If the reconstruction pulse used is  
t
p(t) = ∆
2Ts
specify an equalizer filter E(f ) to recover g(t).

6. For the source

 signalg(t) shown in Figure
 , sketch and compare
 the interpolated pulse signal g̃(t) for
t 1 t 1 2πt
(a) p(t) = Π Ts − 2 , (b) p(t) = ∆ Ts − 2 , (c) p(t) = sin Ts [u(t) − u (t − Ts /2)] and (d) p(t) =
 
∆ 2Tt s in the time domain.

7. Consider a band-limited signal g1 (t) whose Fourier transform is

G1 (f ) = 5 · ∆(f /800)

(a) If g1 (t) is uniformly sampled at the rate of fs = 400 Hz, show the resulting spectrum of the ideally
sampled signal.
 (b) If we attempt to reconstruct g1 (t) from the samples in Part (a), what will be the recovered analog
signal in both time and frequency domains?
(c) Determine another analog signal G2 (f ) in frequency domain such that its samples at fs = 400 Hz
will lead to the same spectrum after sampling as in Part (a).
(d) Confirm the results of (c) by comparing the two sample sequences in time domain.
8. Consider a band-limited signal g(t) whose Fourier transform is
G(f ) = 5 · ∆(f /800)

The sampling frequency fs = 700 Hz will be used.

(a) Show the resulting spectrum of the ideally sampled signal without using any anti-aliasing LPF before
sampling.
(b) Applying ideal interpolation filter to recover g(t) from the samples in Part (a), find the energy of
the resulting error signal g(t) − ge(t). Hint: One may apply Parseval’s Theorem.
(c) Show the resulting spectrum of the ideally sampled signal if an ideal antialiasing LPF of bandwidth
B Hz is used before sampling. What would be an appropriate choice of B ?
(d) Applying ideal interpolation filter to recover g(t) from the samples in Part (c), find the energy of
the resulting error signal g(t) − ge(t) and compare with the result in Part (b).
9. A zero-order-hold circuit (Fig.) is often used to reconstruct a signal g(t) from its samples.
(a) Find the unit impulse response of this circuit.
(b) Find and sketch the transfer function H(f ).
(c) Show that when a sampled signal ḡ(t) is applied at the input of this circuit, the output is a staircase
approximation of g(t). The sampling interval is Ts .

10. (a) A first-order-hold circuit can also be used to reconstruct a signal g(t) from its samples. The impulse
response of this circuit is  
t 1
h(t) = ∆ −
2Ts 2
where TS is the sampling interval. Consider a typical sampled signal ḡ(t) and show that this circuit
performs the linear interpolation. In other words, the filter output consists of sample tops connected
by straight-line segments.
(b) Determine the transfer function of this filter and its amplitude response, and compare it with the
ideal filter required for signal reconstruction.
11. Prove that a signal cannot be simultaneously time-limited and bandlimited. Hint: Show that the contrary
assumption leads to contradiction. Assume a signal simultaneously time-limited and bandlimited so that
G(f ) = 0 for |f | > B. In this case, G(f ) = G(f )Π (f /2B ′ ) for B ′ > B. This means that g(t) is equal to
g(t) ∗ 2B ′ sinc (2πB ′ t). Show that the latter cannot be time-limited.

12. In a non-ideal sampler, the following pulse

 
t
qa (t) = ∆
TS /2
is used as the time-averaging filter impulse response. The sampling rate fs is selected to be higher than
the Nyquist frequency. Design a reconstructed system diagram to recover the original analog signal.
Determine all the necessary filter responses.

13. Repeat 5.1 − 12 when the time-averaging pulse is

 
2πt
qb (t) = sin · [u(t) − u (t − Ts )] .
Ts

14. A signal g(t) bandlimited to B Hz is sampled by using a periodic pulse train pTs (t) made up of a
rectangular pulse of width Tq second (centered at the origin) repeating at the Nyquist rate ( 2B pulses
per second).

(a) Design a reconstructed system diagram to recover the original analog signal when Tq = (8B)−1 .
Determine all the required filter responses in the reconstruction system.
(b) Repeat part (a) if we increase Tq to (4B)−1 and (2B)−1 , respectively. Discuss the effect of different
Tq selections.

15. A compact disc (CD) records audio signals digitally by using PCM. Let the audio signal bandwidth be
15kHz.

(a) If the Nyquist samples are uniformly quantized into L = 65, 536 levels and then binary-coded,
determine the number of binary digits required to encode a sample.
(b) If the audio signal has average power of 0.1 W and peak voltage of 1 V, find the resulting ratio of
signal to quantization noise (SNR) of the uniform quantizer output in part (a).
(c) Determine the number of binary digits per second (bit/s) required to encode the audio signal.
(d) For practical reasons discussed in the text, signals are sampled at a rate well above the Nyquist
rate. Practical CDs use 44,100 samples per second. If L = 65, 536, determine the number of bits per
second required to encode the signal and the minimum bandwidth required to transmit the encoded
signal.

16. A television signal (video plus audio) has a bandwidth of 4.5MHz. This signal is sampled, quantized,
and binary-coded to obtain a PCM signal.

(a) Determine the sampling rate if the signal is to be sampled at a rate 2/9 above the Nyquist rate.
(b) If the samples are quantized into 1024 levels, determine the number of binary pulses required to
encode each sample.
(c) Determine the binary pulse rate (bits per second) of the binary-coded signal.
(d) Find the minimum bandwidth required to transmit this PCM signal.

17. In a satellite radio system, 500 stations of stereo quality are to be multiplexed in one data stream. For
each station, two (left and right) signal channels each of bandwidth 15, 000 Hz are sampled, quantized,
and binary-coded into PCM signals.
 (a) If the maximum acceptable quantization error in sample amplitudes is 1% of the peak signal voltage,
find the minimum number of bits needed for a uniform quantizer.
(b) If the sampling rate must be 8% higher than the Nyquist rate, find the minimum bit rate of the
multiplexed data stream based on the quantizer of part (a).
(c) If 2% more bits are added to the multiplexed data for error protection and synchronization, determine
the minimum bandwidth needed to transmit the final data stream to receivers.

18. A message signal m(t) is normalized to peak voltages of ±1 V. The average message power equals
120 mW. To transmit this signal by binary PCM without compression, uniform quantization is adopted.
To achieve a required SNR of at least 36 dB, determine the minimum number of bits required to code
the uniform quantizer. Determine the actual SNR obtained with this newly designed uniform quantizer.

19. A signal bandlimited to 1MHz is sampled at a rate 30% above the Nyquist rate and quantized into 256
levels using a µ-law quantizer with µ = 255.

(a) Determine the approximate SNR.

(b) Assume that the SNR (the received signal quality) found in part (a) was unsatisfactory. It must be
increased at least by 3 dB. Would you be able to obtain the desired SNR without increasing the
transmission bandwidth (or data rate) if a sampling rate 8% above the Nyquist rate were found to
be adequate? If so, explain how. What is the maximum SNR that can be realized in this way?

20. To digitize an analog signal m(t), it is found that its peak value mp = 8 V, whereas its average power
can vary between 0.2 W to 16 W. L = 256 quantization levels are used.

(a) Determine the SNR range of uniform quantizer output.

(b) Determine the SNR range of nonuniform quantizer output using µ-law compandor of µ = 100 and
µ = 255, respectively.

21. For a PCM signal, determine the minimum necessary L if the compression parameter is µ = 100 and
the minimum SNR required is 42 dB. Recall that L = 2n is required for a binary PCM. Determine the
output SNR for this value of L.

22. Five sensor output signals, each of bandwidth 240 Hz, are to be transmitted simultaneously by binary
PCM. The signals must be sampled at least 20% above the Nyquist rate. Framing and synchronizing
requires an additional 0.5% extra bits. The PCM encoder of previous problem is used to convert these
signals before they are time-multiplexed into a single data stream. Determine the minimum possible data
rate (bits per second) that must be transmitted, and the minimum bandwidth required to transmit the
multiplex signal.

23. Consider a simple DPCM encoder in which N = 1 is used for m(t) = Am cos (ωm t + θm ). The sampling
interval is Ts such that m[k] = m (kTs ) with θm = 0.5ωm Ts . The first-order estimator is formed by

b q [k] = m[k − 1]
m

with prediction error

d[k] = m[k] − m
b q [k]
= Am [cos (kωm Ts + θm ) − cos (kωm Ts + θm − ωm Ts )]

(a) Determine the peak value of d[k].

(b) Evaluate the amount of SNR improvement in dB that can be achieved by this DPCM over a standard
PCM.
 Hint: Let x = kωm Ts + θm . Define a function f (x) = Am [cos(x) − cos (x − ωm Ts )]. Then show that the
maximum values of f (x) can be obtained at x = π/2 + θm + ℓπ, ℓ = 0, ±1, ±2, · · · .

24. A DM system has input message signal

m(t) = 50e−200t cos 1000πt · u(t)

(a) Determine the minimum step size E necessary to avoid slope overload.
(b) Calculate the minimum average quantization noise power based on part (a).

25. Consider a message signal as input to the DM system:

√
m(t) = 3 cos 890πt − 0.7 sin(1000 3πt)

(a) Determine the minimum step size E necessary to avoid DM slope overload.
(b) Calculate the minimum average quantization noise power based on part (a).

26. Consider a signal g(t) = 2 sin 2πt/T is sampled at 4 times of Nyquist frequency.

(a) Determine the sample position for one period (consider there are no sample at t = 0 and t = T )
(b) Find sample values for one period.
(c) Draw the PAM signal considering pulse duration 10% of sampling time.
(d) Draw the PPM signal considering pulse duration 10% of sampling time.
(e) Draw the PWM signal.

27. For a random data sequence of 1011001010001101, sketch the baseband signal waveforms if the line code
is (i) polar RZ, NRZ; (ii) on/off; (iii) bipolar NRZ RZ; (iv) Manchester (v) Polar Quaternary NRZ.