American University in Cairo (AUC)

Electronics and Communications Engineering Department

ECNG 420/4301 Fundamentals of Communications I

Problem Set 4
Spring 2025

Analog to Digital Conversion

Issued: Saturday, April 4th, 2025 Due: Saturday, April 11th, 2025

The homework problems are compiled/adapted from the following references:

ˆ B. Lathi, D. Zing “Modern digital and analog communication systems”, Oxford University
Press, 5th ed., 2019.
ˆ J. Proakis, M. Salehi, “Fundamentals of Communication Systems”, Pearson, 2nd ed., 2014.
ˆ M. Moher, and S. Haykin. Communication Systems. 5th edition, Wiley, 2009.

Problem 4.1
Specify the minimum sampling frequency for the following signals:

1. x1 (t) = sinc(200t).

2. x2 (t) = sinc(200t) + sinc2 (200t).

3. x3 (t) = sinc(200t)sinc(500t).

4. x4 (t) = x1 (t) cos(80000πt)

Problem 4.2
For the message signal m(t) defined by:

m(t) = Am cos(2πfm t)

is represented using PAM pulses. Assuming that the frequency fm = 0.25 Hz, sampling
period Ts = 1s, and Tp = 0.45s.

1. Sketch the spectrum of the PAM signal corresponding to m(t).

2. Using an ideal reconstruction filter, plot the spectrum of the filter output. Compare
this result with the output that would be obtained if there were no distortion (a.k.a.,
aperture effect).

1
Problem 4.3
In this problem, we investigate a variant of the natural sampling discussed in the lectures.
In this variant, the message signal is multiplied by a train of triangular pulse train cT (t)
defined as follows:
∞  
X t − nTs
cT (t) = tri
n=−∞
Tp

For a message signal defined as:

m(t) = sinc(1000t)

1. Calculate the Nyquist rate fN .

1
2. Sketch the spectrum of naturally-sampled signal with fs = 4fN and Tp = 4fs
.

3. Suggest an appropriate receiver for reconstructing m(t) from its natural samples.
Do we need to have an equalizer at the receiver in this case?

Problem 4.4
Twenty-four voice signals are sampled uniformly and then time-division multiplexed. The
sampling operation uses flat-top samples with 1µs duration. The multiplexing operation
includes provision for synchronization by adding an extra pulse of sufficient amplitude
and also 1µs duration. The highest frequency component of each voice signal is 3.4 KHz.

1. Assuming a sampling rate of 8 KHz, calculate the spacing between successive pulses
of the multiplexed signal.

2. Repeat the calculation for Nyquist-rate sampling.

Problem 4.5
A PCM system using a uniform quantizer followed by a 7-bit binary encoder. The bit
rate of the system is 50 Mb/s.

1. What is the maximum message bandwidth for which the system operates satisfac-
torily?

2. Determine the output signal-to-quantization noise ratio when a full-load sinusoidal

modulating wave of frequency 1 MHz is applied to the input.

2
Problem 4.6
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
KHz are sampled, quantized, and binary-coded into PCM signals.

1. 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 quan-
tizer.

2. 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 1.

3. If 2% more bits are added to the multiplexed data for error protection and syn-
chronization, determine the minimum bandwidth needed to transmit the final data
stream to receivers.

Problem 4.7
A compact disc (CD) records audio signals digitally by using PCM. Let the audio signal
bandwidth be 15 KHz.

1. If the Nyquist samples are uniformly quantized into L = 65536 levels and then
binary-coded, determine the number of binary digits required to encode a sample.

2. 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 ratio (SQNR) of the uniform quantizer
output in part 1.

3. Calculate the minimum bit rate to encode the audio signal.

4. For practical reasons, signals are sampled at a rate well above the Nyquist rate.
Practical CDs use 44100 samples per second. If L = 65536, determine the bit rate
required to encode the signal and the minimum bandwidth required to transmit
the encoded signal.

Problem 4.8
The message signal:

m(t) = 6 sin(2πt), volts

is transmitted using a 4-bit binary PCM system. The quantizer is of the mid-rise type,
with a step-size of 1 volt. Assume a sampling rate of four samples per second, with
samples taken at t = ±1/8, ±3/8, ±5/8, ... seconds. Sketch the resulting PCM wave
for one complete cycle of the input using polar RZ line code. Repeat the problem for
mid-tread quantizer with the same step size.

3
Problem 4.9
In a PCM system, the output of the quantizer is the digital sequence dn = 10110010. The
sequence is line coded using waveforms of amplitude A = 2 and bit duration Tb = 1µs:

ˆ Polar NRZ.

ˆ Unipolar RZ.

ˆ Machester coding.

ˆ Bipolar RZ (Assume first 1 has amplitude of +2).

ˆ Differential encoding (Assume a0 = 1).

For these line codes:

1. Sketch the waveforms corresponding to the digital sequence {dn }8n=1 .

2. From the list, choose line code(s) that do not have DC component.

3. From the list, choose line code(s) that exhibit self-synchronization.

4. From the list, choose line code(s) that consume the least bandwidth.

Problem 4.10
Consider the message signal m(t) defined by the hyperbolic tangent function:

m(t) = A tanh(βt)

where A, and β are fixed parameters. The message m(t) is sampled with a sample time
Ts and converted to digital using delta modulation (DM).

1. Find the minimum step size of DM to avoid slope-overload distortion.

2. Calculate the minimum average quantization noise power.

Problem 4.11
An analog signal is sampled with Ts = 2µs. The first 10 samples of the signal is as
follows:

m(nTs ) = [2.2 3.87 6.146 7.46 7.356 6.35 4.846 3.125 2.232 2.872]

The samples are converted to digital with the following methods:

ˆ Delta modulation with step size ∆ = 1.

ˆ DPCM with 2-bit mid-rise quantizer with step size ∆ = 1 and a message predictor
m̂(nTs ) = mq ((n − 1)Ts ).

4
 ˆ DPCM with 2-bit mid-rise quantizer with step size ∆ = 1 and a message predictor
m̂(nTs ) = 2mq ((n − 1)Ts ) − mq ((n − 2)Ts ).

For each case,

1. Find the output binary sequence.

2. Calculate the bit rate.

3. Sketch the transmitter and the receiver block diagrams showing the details of the
message predictor.

Compare visually, which method is better. Explain your reasons.

Portfolio Questions (Bonus)

The following questions are not a part of the assignments and will not be graded as an
assignment. All Portfolio questions will be collected and graded once at the end of the
semester as a BONUS.

1. Using sampling theorem, prove that: for fs > 0

∞
X
sinc(fs t − n) = 1
n=−∞

2. Repeat the example in Slide Set 5 Slide 28 for PAM signal with Tp = Ts (a.k.a.,
staircase representation).

3. A TDM-PCM system operates with voice signals with bandwidth of 5 KHz. The
number of levels of the quantizer is 256 levels. If the total bit rate of the system
is 1.54 Mb/s. Find the maximum number of voice signals that can be multiplexed
together using this TDM-PCM system.

4. Search for expressions for the SQNR with µ-law companders. Does the overall
SQNR enhances with increasing µ?

5. How to decode the original bit sequence {dn } from the differentially-encoded se-
quence of bits {an }. Write an equation describing the decoder.

6. Which line code has lower bit error rate for the same amplitude A, polar NRZ, or
unipolar RZ. Explain intuitively why.