DHARMSINH DESAI UNIVERSITY, NADIAD
FACULTY OF TECHNOLOGY
B.TECH - Semester - VI
SUBJECT: (EC611) DIGITAL SIGNAL PROCESSING
Examination : Third Sessional Exam Seat No. :
Date : 24/03/2015 Day : Tuesday
Time : 12:45 to 2:00 pm Max. Marks : 36
INSTRUCTIONS:
1. Figures to the right indicate maximum marks for that question.
2. The symbols used carry their usual meanings.
3. Assume suitable data, if required & mention them clearly.
4. Draw neat sketches wherever necessary.
5. Bold letter in sequence represents n=0 position of sequence.
Q.1 Do as directed. [6]
(a) Convert an analog filter with transfer function H(s) = 4 into digital filter using [2]
( s 2)( s 3)
Matched Z-transform method.(Assume ts = 0.5sec.).
(b) Compare various conversion methods from Analog transfer function to digital filter transfer [2]
function.
(c) What are the advantages of FIR filter? Why? [2]
Q.2 Answer the following. (Any two.) [12]
(a) (i) Design a 4th order Butterworth-type IIR low-pass digital filter using bilinear [4]
transformation with 3dB cut-off at one sixteenth of the sampling frequency f s.
Transfer Function of low pass filter at 1 rad/sec is given as
(ii) Why Bilinear Transformation is more preferred over other conversion methods? [2]
(b) Design an ideal highpass FIR filter having cutoff frequency of 1.5kHz and sampling [6]
frequency of 5kHz with 7 samples using hann window. Plot final frequency response of the
system.
(c) If H(z) has zeros a z1 = 1/√2 + j/√2, z2 = 2. Determine the lowest degree H(z) that has a linear [6]
phase.
Q.3 Do as directed.
(a) If N = 64, Decimation in frequency FFT algorithm is used. X(25) is available at which [1]
location?
(b) What is required in DSP processor to support VLIW. [1]
(c) Represent -25/64 in sign magnitude format assuming 4 bit register length whose MSB [2]
represents sign. Compute quantization error.
(d) Differentiate Von Neuman and Harvard architecture. [2]
Q.4 (a) Discuss Adaptive filters and any two applications. [3]
(b) Describe long filtering using DFT. [6]
Apply Overlap save method and find out response of the system h(n)=[1 1 1] and
x(n)=[1 2 2 3 4 5].
(c) Find the DFT of x(n) = {2, -1, 4 3} using Decimation in Time FFT algorithm. [3]
OR
Q.4 (a) Discuss Multi Rate Signal Processing techniques. [4]
Explain the procedure to find fractional delay x(n-2/3).
(b) Discuss Divide and Conquer approach to reduce no. of computations in finding DFT. [4]
(c) Explain DSP algorithm requirements. [4]
Page 1 of 1