PES INSTITUTE OF TECHNOLOGY BANGALORE SOUTH

CAMPUS

(Formerly, PES SCHOOL OF ENGINEERING, Hosur

Road, 1 KM BEFORE ELECTRONIC CITY, BANGALORE)

QUESTION BANK

DIGITAL SIGNAL PROCESSING (15EC52)

Module 1 (INTRODUCTION TO DSP, DFT, DFT PROPERTIES)

1. What is Digital Signal Processing (DSP)? What are the basic

elements of a DSP system?

2. Compare Analog Signal Processing and Digital Signal Processing.

3. What are the limitations of DSP?

4. Give examples for DSP applications.

5. Define discrete time signals and classify them.

6. Explain how an analog signal can be converted to a discrete time

signal.

7. What is aliasing? Explain with the help of an example.

8. What is Linear Convolution? Convolve the sequences x(n) =

and h(n) = .

9. Give the mathematical expressions for the following: (a) Fourier

Series; (b) Inverse Fourier Series; (c) Fourier Transform; (d) Inverse
Fourier Transform; (e) Discrete Time Fourier Series; (f) Inverse
Discrete Time Fourier Series; (g) Discrete Time Fourier Transform; (h)
Inverse Discrete Time Fourier Transform;

10. Explain the need for Discrete Fourier Transform (DFT). Define
DFT for a sequence x(n).

1
11. Define Z Transform for a discrete time signal x(n). Explain the
significance of ROC in Z Transform.

12. Derive the expression for DFT from DTFT expression.

13. Show that DFT and IDFT form a consistent Discrete Fourier
Transform pairs.

14. Let X(k) denote the N point DFT of an N point sequence x(n).If
the DFT X(k) is computed to obtain a sequence x1 (n) , determine x1 (n) in
terms of x(n).

15. Establish the relation between DFT and ZT.

16. Establish the relation between DFT and DFS.

17. What are twiddle factors in DFT?

18. Explain the methods for computation of DFT in linear filtering of

long duration sequences with appropriate diagrams.

19. Compute the DFT of the sequence x(n) = {0, 1, 2, 3}.

20. Calculate the 8 point DFT of the sequence x(n) = {1, 1, 1, 1}.

21. Compute the DFT of the following standard signals: (a) x(n) =
δ(n); (b) x(n) = a n for 0  n  N-1 ;

22. State and prove the following properties of DFT: (a) Periodicity;
(b) Linearity; (c) Time shifting property; (d) Time reversal of a
sequence; (e) Complex conjugate property; (f) Multiplication of 2
DFT’s – Circular Convolution property; (g) Circular Correlation
property; (h) Multiplication of two sequences; (i) Parseval’s Theorem;
(j) Circular Time Shift property; (k) Circular Frequency Shift property;

23. State the periodic convolution of two discrete sequences.

24. Distinguish between linear convolution and circular convolution

of two sequences.

25. Give the steps to get the result of linear convolution from the
method of circular convolution.

26. Explain the meaning of ‘frequency resolution’ of the DFT.

2
27. Convolve the following sequences using (a) overlap – add
method; (b) overlap – save method; for x(n) = {1, -1, 2, 1, 2, -1, 1,
3, 1} and h(n) = {1, 2, 1}.

1 
 ,0n2
28. Determine the DFT of the sequence h(n)=  3 .
0, otherwise 

29. Compute the inverse DFT for the sequence X(k) = e-k for k = 16.

30. For the 8 sample sequence x(n) = {1, 2, 3, 5, 5, 3, 2, 1}, the

first 5 DFT coefficients are {22, -7.5355 – j 3.1213, 1 + j, -0.4645 – j
1.123, 0}. Determine the remaining 3 DFT coefficients.

Module 2

1. Give the number of complex additions and complex multiplications

required for direct computation of N point DFT.

2. What is the need for Fast Fourier Transform (FFT) algorithm?

3. Why is FFT called so?

4. State the computational requirements of FFT.

5. Compare the number of multiplications required to compute the

DFT of a 64 point sequence using direct computation and that using
FFT.

6. What is DIT-FFT algorithm?

7. What is DIF-FFT algorithm?

8. Which FFT algorithm procedure is the lowest possible level of DFT

decomposition?

9. Give the computation efficiency of FFT over DFT.

10. Compute the FFT of the sequence x(n)=n 2  1 , where, N = 8 using

DIT algorithm.

11. Draw the flow graph of an 8 point DIF-FFT and explain.

3
12. Discuss the computational efficiency of radix 2 FFT algorithm.

13. Explain how you would use FFT algorithm to compute IDFT.

14. Develop the decimation-in-time and decimation-in-frequency FFT

algorithms for decomposing DFT for N = 3.3.3 and obtain the
corresponding signal flow graphs.

15. Determine the DFT of the following sequence using DIF-FFT

algorithm: x1 (n) = {1, 1, 1, 0, 0, 1, 1, 1}. Using the DFT of x1 (n) find
the DFT of the sequence x 2 (n) = {1, 1, 1, 1, 1, 0, 0, 1}.

16. What are the advantages of fast (sectioned) convolution?

Module 3 (ANALOG FILTER DESIGN)

1. Explain the frequency response characteristics of Butterworth

filters.

2. What are frequency transformations? Why is it required?

1
3. Given | Ha ( j)|2  , determine the analog filter system function
1  646
Ha (s) .

4. Design an analog filter with maximally flat response in the

passband and an acceptable attenuation of -2 dB at 20 rad/sec. The
attenuation in stopband should be more than 10 dB beyond 30
rad/sec.

5. Design a low pass 1 rad/sec bandwidth Butterworth filter with

the following characteristics: (i) Acceptable passband ripple of 2 dB;
(ii) Cutoff radian frequency of 1 rad/sec; (iii) Stopband attenuation of
20 dB or greater beyond 1.3 rad/sec.

6. Using analog frequency transformation design an analog

Butterworth filter with the following specifications: (i)Passband ripple:
1 dB for 0    10 rad/sec ; (ii) Stopband ripple: - 60 dB for Ω  50 rad/sec .

7. Using analog frequency transformations design a highpass

Butterworth filter of 4th order for cutoff frequency of 50 Hz.

4
8. Using analog frequency transformations design a second order
bandpass Butterworth filter with passband of 200 Hz to 300 Hz.

9. Determine the normalized low pass Butterworth analog poles for

N = 10.
10. A difference equation describing a filter is given by y(n) – ¾ y(n-
1) + 1/8 y(n-2) = x(n) + ½ x(n-1). Draw direct form I and direct form
II structures.
11. Realize the following system function in parallel form
2 7 1
1 - z -1 1 + z -1 - z -2
H(z) = 3 . 4 2
7 3 -2 1
1 - z -1  z 1 - z -1 + z -2
8 32 2
12. Realize the following system function in cascade form
(1 - z-1 )3
H(z) =
 1 -1  1 -1 
1 - z 1 - z 
 2  8 
13. Obtain the parallel form realization of the following system
(3 + z-1 )
function H(z) 
 1  1 
3 1 - z -1 1 - z -1 
 2  4 
14. An analog filter has the following system function. Convert this
filter into a digital filter using impulse invariant technique:
1
H(s) =
(s + 0.1)2 + 9

15. Use the backward difference for the derivative to convert the
analog low pass filter with the following system function, using impulse
invariant transformation H(s) = 1/(s+2).

16. Transform the analog filter with the transfer function shown
below into a digital filter, using backward difference for the derivative:
H(s) = 1/(s+2)(s+3).

17. Convert the analog filter to a digital filter whose system function
36
is H(s) = . The digital filter should have a resonant
(s + 0.1)2 + 36
frequency of ωr = 0.2 π . Use bilinear transformation.

18. Convert the analog filter to a digital filter whose system function
1
is H(s) = using bilinear transformation.
(s + 2)2 (s +1)

5
19. Design and realize a digital LPF using bilinear transformation to
satisfy the following requirements: (i) monotonic passband and
stopband; (ii) – 3 dB cut off frequency at 0.6π radians; (iii) magnitude
down at 16 dB at 0.75π radians

20. Design a digital Butterworth filter to meet the following

constraints using (i) Bilinear transformation; (ii) Impulse invariant
method
0.8  |H(e jw ) |  1, 0  w  0.2π
|H(e jw ) |  0.2, 0.26π  w  π

Module 4

1. What is the basic difference equation form for FIR filters?

2. Explain Linear phase FIR structures.

3. Realize a linear phase FIR filter with the following impulse response.
Give necessary equations: h(n) = δ(n) + ½ δ(n-1) – ¼ δ(n-2) + δ(n-
4) + ½ δ(n-3).

4. Realize the following system function by linear phase FIR structure:

H(z) = 2/3 z + 1 + 2/3 z -1 .

5. Realize the following system function in cascade form:

3 17 -2 3 -3
H(z) = 1 + z -1  z + z + z -4 .
4 8 4

6. Explain the frequency sampling structure for FIR systems.

4z 2
7. H(z) = 3 +  . (i) Does this H(z) represent a FIR or IIR
z - 0.5 z - 0.25
filter? Why?; (ii) Give a difference equation realization of this system
using direct form I; (iii) Draw the block diagram for the direct form II
canonic realization, and give the governing equations for
implementation.

8. Explain the method for designing FIR filters using the window
technique.

6
9. Explain Gibb’s phenomenon with respect to window technique for
designing FIR filters.

10. Name the different types of window functions? How are they
defined (mathematically).

11. The desired frequency response of a low pass filter is

1,   / 2     / 2 
H d (e j )    . Determine hd (n) . Also determine h(n) using
0,  / 2 |  |  
the symmetric rectangular window with window length = 7.

12. The desired frequency response of a low pass filter is

e j 3 ,  3 / 4    3 / 4 
H d (e j )   j
 . Determine H (e ) for M = 7 using a
0, 3 / 4 |  |  
rectangular window.

13. Design a bandpass filter which approximates the ideal filter with
cut off frequencies at 0.2 rad/sec and 0.3 rad/sec. The filter order is M
= 7. Use the hanning window function.

14. Explain Frequency Sampling method of designing an FIR filter.

15. A low pass filter has the desired response given by

j e j 8 , 0     / 2 
H d (e )    . Using the frequency sampling (Type I)
0,  / 2 |  |  
technique, determine the filter coefficients. The length of the filter is M
= 17.

16. Obtain a general expression for the frequency response of linear

phase FIR filters.