School of Engineering
Department of Computer and Communication Engineering
Fall 2018
Course: EENG527 – Digital Signal Processing
Instructors: Dr. Ammar El Falou, Dr. Ayman Khalil, Dr. Hussein Kobeissi, Dr. Hussein Nasrallah,
Dr. Mohamad Husseini, Dr. Walid Kaakour, Dr. Hussein Hijazi
Midterm Examination
Date: 19 November 2018
Time: 120 mins
Student Name: _________________________ Section:_________________
There are four Questions in the booklet each has several parts, please answer all parts of the 4 questions to
the best of your ability.
Marking Scheme:
Questions Weight Mark
Question 1 35 points
Question 2 20 points
Question 3 30 points
Question 4 15 points
Total 100 points
1. This booklet contains 14 pages including this one. Make sure you have all these pages.
2. Closed Book Examination.
3. Do not take the staple out. The exam booklet must remain intact
4. The cheating penalty will be “F” in the exam
Good Luck
1/10
Question 1: (35pts)
1) List three advantages for digital processing over analog processing and give one practical limitation
for digital implementation. (5pts)
Digital system can be simply reprogrammed for other applications/ported to different hardware /
duplicated.
DSP provides better control of accuracy requirements (ADC and digital processor)
More sophisticated signal processing algorithms can be implemented in DSP
One practical limitation: Speed of operation of ADC and digital signal processors Extremely wide
bandwidths signals require fast-sampling-rate ADC and fast digital signal processors.
2) Give three real applications for the DSP block. (3pts)
Digital Filtering, Spectrum analyzer, Electrocardiography, Speech Coding and Compression, Compact-Disc
Recording, Digital Photo Image Enhancement, Digital Communication
3) A digital communication link carries binary-coded words representing samples of an input signal:
( ) ( ) ( )
The link is operated at 5,000 bits/s and each input is quantized into 1024 different voltage levels.
a. Sketch the bloc diagram of the analog to digital converter (ADC) and describe each bloc and its
parameters. (5pts)
The bloc diagram of the ADC is:
Sampler: converts a continuous-time signal into a discrete-time signal. It is characterized by a
sampling rate Fs. Fs = number of samples per second.
Quantizer: converts a continuous valued signal into a discrete-valued signal belonging to an
Alphabet. It is characterized by the step resolution Δ and number of levels L.
Coder: It codes the quantized samples into a binary stream. It is characterized by the number
of bits per sample b.
2/10
b. What are the sampling and folding frequencies? (3pts)
Number of bits/sample = ( ) bits
samples/sec = 500 Hz
=500/2=250 Hz
c. The analog signal ( ) is given in the figure below for two periods. Sketch on the same figure the
sampled signal ( ). (5pts)
T = sampling period = 1/500 = 2 ms
The period of ( ) is equal to LCM(0.005;0.01) = 0.01 = 10 ms.
3/10
d. What is the Nyquist rate for the signal ( )? Is the considered sampling frequency complying with
the Nyquist theorem? (3pts)
200 Hz 2 = 400 Hz. Yes,
e. What is the resolution ? (2pts)
( )
f. Compute the SQNR in dB. (2pts)
SQNR = 1.76 + 6.02*b = 61.96 dB.
g. Sketch on the same figure the reconstructed signal using a zero-holder interpolator and linear
interpolator. Is the obtained signal similar to the original one? (5pts)
As is very small, we can assume that the quantized signal and the sampled signal are equivalent.
The obtained figure is:
4/10
It is clear that the obtained signal is not similar to the original one. The linear interpolator is better
than the zero-holder.
h. What do you suggest in order to enhance the quality of the obtained signal with zero-holder of linear
interpolator? What are the consequences on our system? (2pts)
In order to obtain a good signal, we should increase the sampling frequency to higher value e.g.,
. In this case, we should increase the transmission rate to *b = 10 Kbits/second. Note
that the augmentation of the number of quantization bits will not improve the quality if we use zero-
holder or linear interpolator.
5/10
Question 2: (20 pts)
Consider the following system ( ) ( ) ( )
1) Give the output for the input: (5pts)
( ) { }
( ) { }
( ) { }
( ) { }
2) Examine the system with respect to the properties below: (15pts)
a- static or dynamic
output at time n does not depend on the input at time n dynamic
b- linear or nonlinear
( ) ( ) ( )
( ) ( ) ( )
( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )
The system is linear
c- causal or noncausal
Causal: The output at instant n depends only at the present and past inputs
d- stable or unstable
Consider a bouded input |x(n)| < M for all n
-M < x(-n2) < M and -2M < 2x(n-1) < 2M and -3M <y(n) < 3M Output is bounded system is stable
e- time variant or time invariant
( ) ( ( ) ) ( )
( ) ( ) ( )
The system is time-variant
6/10
Question 3: (30 pts)
Consider an LTI discrete-time system described by a difference equation:
( ) ( ) ( ) ( ) ( ) for n ≥ 0
In the sequel, we assume the system is initially relaxed. Designate by h(n) the impulse response of the
system.
1- Decompose the system into two subsystems in cascade: one non-recursive and another one purely
recursive. Give the input-output relation for each subsytem. (4pts)
( ) ( ) ( )
( ) ( ) ( ) ( )
2- Determine the direct form II realizations of the system. (6pts)
3- Explain how we can obtain h(n). (3pts)
Assuming system is relaxed and x(n) = δ(n) y(n) = h(n)
4- Compute recursively the terms h(0) and h(1) (using the difference equation). (4pts)
( ) ( ) ( ) ( ) ( ) = -1/2
( ) ( ) ( ) ( ) ( ) = -3/4
7/10
5- Determine the expression of h(n) in a compact form by solving the difference equation. (5pts)
Solving for the homogeneous solution, we get 2r2 -3r +1 = 0 r1 = 1 and r2 = 1/2
( ) ( ) [ ( ) ] ( )
( )
( )
½ C2 = ¼ C2 = ½ and C1 = -1 ( ) [ ( ) ] ( )
6- Deduce the stability of the system. Is this an FIR or an IIR system? (4pts)
The system is unstable since one of roots is .
( ) [ ( ) ] ( ) system is IIR
7- Consider now the input x(n) = {0 1 2}. Find its output y(n). (4pts)
x(n) = δ(n-1) + 2 δ(n-2) y(n) = h(n-1) + 2 h(n-2) (since system is relaxed)
Question 4: (15 pts)
In some applications such as radar signal processing, seismic analysis, and signal enhancement, an
information-bearing signal is obscured by some form of noise. The correlation functions are used to estimate
certain signal characteristics from the corrupted signal.
In this exercise, we consider the application of correlation functions to radar system. The underlying
principle of radar is that an electromagnetic pulse transmitted at a target will be reflected by the target and
will consequently return to the sender with a delay proportional to the distance to the target. In what follows,
you will aim to estimate such a delay in order to locate the distance of the target.
8/10
Let x(n) = ( ) - ( ) be the original transmitted signal and y(n) = { 0.1 1.3 -0.2 0.9 -2.1} the reflected
noisy signal.
D: round-trip delay
α: attenuation
w(n): additive noise
1) Determine if these signals x(n) and y(n) are energy-type or power-type. (2pts)
These signals are finite duration so energy signals
2) Give ( ) ( ) and ( ) ( ). (3pts)
( ) ( ) = { 0.1 1.3 -0.2 0.9 -2.1}
( ) ( ) = { 0.1 1.3 -0.2 0.9 -2.1}
3) Deduce that the correlation function ( ) between y(n) and x(n) is given by: (3pts)
( ) { }
Hint: ( ) ( ) ( )
( ) ( ) ( )= ( ) [ ( ) ( )] { }
9/10
4) Explain how we can estimate the round-trip delay D from the correlation function. (4pts)
The maximum value of the correlation ryx(k) is obtained at k = D (normalized by Ts).
So D = 3 delay in second = 3*Ts
5) If the sampling rate Fs = 1MHz, determine the distance d between the radar and the target. (3pts)
d = C * D/2 * Ts ( D*Ts : round trip delay in sec)
= 3* 108 * 3/2 * 10-6 = 450 m
10/10