A 060008c301122003
PTA
w
RegNo.: Name:
APJ ABDUL KALAfI TECHNOLOGICAL I]NTVE
Fifth Semester B.Tech Degree (S,FE) ExaminationJanuary 2022
Qffiff)i'
Course Code: EC301
Course Name: DIGITAL SIGNAL PROCESSING
Max:,Marks: 100 Duration: 3 Hours
PART A
Answer any twofalt questions, eoch canies 15 marks.
Marks
l\ a) write down the equation forN point DFT and explain each term (4)
rl
b) Find N point DFT of unit impulse d(n) (4)
c) Find 4 point DFT of x(n) :ll, -2,3,21- Q)
2 a) Find 4 point circular convolution of x(n): h(n): [l,l,l,l] using DFT. (8)
b) Compute 8-point DFT ofthe sequence I I I I 0 0 0 0] using Decimation in Time (7)
FFT algorithm
3 a) Let x(n): {1,0,1,0} and h(n) : {1,2,2,1}. Find 4 point DFTs of these sequences (8)
using a single 4 point DFT.
b) Find the IDFT of X(k): [1,0,1,0] using DIF FFT algorithm. (7)
PART B
Answer any twofull questions, each carries 15 marks.
4 a) Prove that a symmetric impulse response results in a linear phase response for an (8)
FIR filter with an even filter order N.
t b) Compare main 3 properties of rectangular, Hanning and Hamming window (3)
functions.
c) How filter order is selected in the window-based method of FIR filter design. (4)
5 a) Explain impulse invariant mapping. List the main limitations of impulse invariant (5)
mapping.
b) Design a Digital Butterworth filterto satisff the constraints 0.9 < lH(ar)l S (10)
1; 0 < a 3 0.5tt. lH(ar)l < 0.2;O.75tr < a 32. Usp bilinear transformation.
Assume T=1 s.
6 a) Design a Linear phase LPF with a cut offfrequency of 0.5n rad/s using frequency (10)
sampling. Take N: 13, use type I design.
b) Derive the mapping function from s to z in Bilinear transformation. Explain (5)
frequency warping.
Page 1 of 2
ffi000Ec301t22003
:" ':: '" ' PARTC
Answer any twofull quations, each corries 20 mnrks.
7 a))R*li,"thesysfemgivenbyH(z)=ffiincascadeformandpara|lel(l0)
i+g.62-1+o.6 z-2 .
form.
b) Explain transposed form of filter structure. Realise the system y(n) - bsx(n) + (10)
b#(n - L) * b2x(n - - afl(n- 1) - azt(n-
2) Z) in transposed form.
8 a) Explain decimation and interpolation. (6)
b) Explain effects of coefficient quantisation in FIR and IIR filters. (7)
, c) Why lowpass filtering is performed before down sampling? How will you decide (7)
q. the cutofffrequency ofthe low pass filter?
.',
'ia 9 a) Explain the architecture ofTMS32rc,6TXXDSP processorwith a neat diagram. (10)
b) Realise the transfer function H(z)= .!Y'l= ,= in Direct form I and 2. (10)
=
l-l.Sz-r+o-92
Compare the number of delay elements, adders and multipliers needed in both
cases.
**!f *
Page2ot2
