DSP EC 2302 APRIL/MAY 2011

 

 B.E./B.Tech. DEGREE EXAMINATION, APRIL/MAY 2011.
                                      Fifth Semester
            Electronics and Communication Engineering 
            EC 2302 — DIGITAL SIGNAL PROCESSING

(Common to PTEC 2302 Digital Signal Processing for B.E. (Part-Time) Electronics and Communication Engineering Fourth Semester – Regulation 2009)
                                                      (Regulation 2008)
Time : Three hours                                                             Maximum : 100 marks

Answer ALL questions.

PART A — (10 × 2 = 20 marks)
1.
State the advantages of FFT over DFTs.
2.
What is meant by bit reversal?
3.
Why do we go for analog approximation to design a digital filter?
4.
Give any two properties of chebyshev filters.
5.
State the properties of FIR filters.
6.
What is meant by Gibbs Phenomenon?
7.
What is meant by fixed point arithmetic? Give example.
8.
Explain the meaning of limit cycle Oscillator.
9.
State the various applications of DSP.
10.
What is echo cancellation?

PART B — (5 × 16 = 80 marks)
11.
(a)  With appropriate diagrams describe 
 (i) overlap-save method    (8) 
 (ii) overlap-add method.    (8)
Or 
(b)  Explain Radix-2 DIF-FFT algorithm. Compare it with DIT-FFT algorithms.    (16) 
12.
(a)  Explain in detail butterworth filter approximation.  (16)
Or 
(b) Explain the bilinear transform method of IR filter design. What is warping effect? Explain the poles and zeros mapping procedure clearly.       (16)
13.
(a)  Realize the system function ( )      
by linear phase
FIR structure.    (16)
Or 
(b) Explain the designing of FIR filters using windows.  (16)
14.
(a)  Explain the quantization process and errors introduced due to quantization.    (16)
Or 
(b) (i)  Explain how reduction of product round-off error is achieved in digital filters   (8) 
 (ii)  Explain the effects of coefficient quantization in FIR filters.  (8)
15.
(a) (i) Explain how various sound effects can be generated with the help of DSP.    (10) 
 (ii)  State the applications of multirate signal processing.  (6)
Or 
(b) (i) Explain how DSP can be used for speech processing.  (8) 
 (ii)  Explain in detail about decimation and interpolation.  (8)

DSP (ANNA UNIVERSITY) MAY/JUNE 2012 QUESTION PAPER

B.E./B.TECH. DEGREE EXAMINATION, MAY/JUNE 2012 

FIFTH SEMESTER

                                 ELECTRONICS AND COMMUNICATION ENGINEERING 
                                     EC 2302 / EC52 – DIGITAL SIGNAL PROCESSING
                                                              (REGULATION 2008) 

TIME: THREE HOURS                                                                MAX: 100 MARKS 
ANSWER ALL THE QUESTIONS 
PART A – (10 X 2 = 20) 
1. WHAT IS TWIDDLE FACTOR? 
2. HOW MANY STAGES OF DECIMATIONS ARE REQUIRED IN THE CASE OF A 64 POINT RADIX 2 DIT FFT ALGORITHM? 
3. WHY IS THE BUTTERWORTH RESPONSE CALLED A MAXIMALLY FLAT RESPONSE? 
4. WHAT IS FREQUENCY WARPING? 
5. WHAT ARE THE FEATURES OF FIR FILTER DESIGN USING THE KAISER’S APPROACH? 
6. DRAW THE DIRECT FORM IMPLEMENTATION OF THE FIR SYSTEM HAVING DIFFERENCE EQUATION. 
     y(n) = x(n) – 2x(n-1) + 3x(n-2) – 10x(n-6) 
7. WHAT ARE LIMIT CYCLE OSCILLATIONS? 
8. WHAT IS DEAD – BAND OF A FILTER? 
9. WHAT IS DECIMATION? 
10. FIND THE EXPRESSION FOR THE FOLLOWING MULTIRATE SYSTEMS.       
   
                                              PART B – (5 X 16=80) 
11. (a) (i) DIFFERENTIATE DFT FROM DTFT.                                                  (4) 
      (ii) COMPUTE AN 8 POINT DFT OF THE SEQUENCE  x(n)   (1,0,1,-1,1,-1,0,1)                                                                                  (12)  
QUESTION PAPPER CODE: 11332
     4          
24
     4          
[OR]  
     (b) (i) PROVE THAT FFT ALGORITHM HELPS IN REDUCING THE NUMBER OF COMPUTATIONS INVOLVED IN DFT COMPUTATION.                         (6)  
           (ii) COMPUTE A 8 POINT DFT OF THE SEQUENCE USING DIT – FFT ALGORITHM                             X(n)   =(1,2,3,2,1,0)                                                                          (10)                                                                   
12. (a) (i) EXPLAIN THE PROCEDURE FOR DESIGNING ANALOG FILTERS USING THE CHEBYSHEV APPROXIMATION.                                                 (6) 
           (ii)  CONVERT THE FOLLOWING ANALOG TRANSFER FUNCTION IN TO DIGITAL USING IMPULSE INVARIANT MAPPING WITH T = 1 sec.      (10) 
[OR]  
      (b)  (i) DESIGN A DIGITAL SECOND ORDER LOW – PASS BUTTERWORTH FILTER WITH CUT-OFF FREQUENCY  2200 Hz USING BILINEAR TRANSFORMATION. SAMPLING RATE IS 8000 Hz.                      (8) 
            (ii) DETERMINE THE CASCADE FORM AND PARALLEL FORM IMPLEMENTATION OF THE SYSTEM GOVERNED BY THE TRANSFER FUNCTION                                      
                                        (1 + Z -1)(1 – 5Z -1 – Z -2)                     H(Z)=         (1 +2Z -1 + Z -2)(1 + Z -1 + Z -2)                                                       (8) 
13. (a) DESIGN AN FIR LOW PASS DIGITAL FILTER BY USING THE FREQUENCY SAMPLING METHOD FOR THE FOLLOWING SPECIFICATIONS 
CUTOFF FREQUENCY = 1500Hz 
SAMPLING FREQUENCY = 15000 Hz 
ORDER OF THE FILTER : N = 10 
FILTER LENGTH REQUIRED L = N  + 1 = 11                                                     (16) 
[OR] 
(b) (i) EXPLAIN WITH NEAT SKETCH THE IMPLEMENTATION OF FIR FILTERS IN THE  
      (1) DIRECT FORM 
      (2) LATTICE FORM                                                                                             (6) 
      (ii) DESIGN A DIGITAL FIR BAND – PASS FILTER WITH LOWER CUT-OFF FREQUENCY 2000 Hz AND UPPER CUT-OFF FREQUENCY 3200 Hz USING HAMMING WINDOW OF LENGTH N = 7. SAMPLING RATE IS 10000 Hz.                                                                                                                             (10)  
14. (a) (i) WHAT IS QUANTIZATION OF ANALOG SIGNALS? DERIVE THE EXPRESSION FOR THE QUANTIZATION ERROR. 
      (ii) EXPLAIN COEFFICIENT QUANTIZATION IN IIR FILTER. 
[OR] 
     (b) (i) HOW TO PREVENT LIMIT CYCLE OSCILLATION ? EXPLAIN. 
          (ii)WHAT IS MEANT BY SIGNAL SCALING? EXPLAIN.
 
15. (a) (i) EXPLAIN SAMPLING RATE CONVERSION BY A RATIONAL FACTOR AND DERIVE INPUT OUTPUT RELATION IN BOTH TIME AND FR EQUENCY DOMAIN.                                                                                   (10) 
            (ii) EXPLAIN THE MULTISTAGE IMPLEMENTATION OF SAMPLING RATE CONVERSION.                                                                                         (6) 
[OR]  
     (b) (i) EXPLAIN THE DESIGN OF NARROW BAND FILTER USING SAMPLING RATE CONVERSION.                                                                       (8) 
          (ii) EXPLAIN THE APPLICATION OF SAMPLING RATE CONVERSION IN SUB – BAND CODING.                                                                                    (8)