B.E ECE.

6TH SEM
ADVANCED DIGITAL SIGNAL PROCESSING
MAX:100
ANSWER THE FOLLOWING:

PART A-(10*2=20)
1. Define statistical variance and covariance.
2. How do you compute the energy of a discrete signal in time and frequency domains?
3. Define sample autocorrelation function. Give the mean value of this estimate.
4. What is the basic principle of Welch method to estimate power spectrum?
5. How do find the ML estimate?
6. Give the basic principle of Levinson recursion.
7. Why are FIR filters widely used for adaptive filters?
8. Express the Widrow- Hoff LMS adaptive algorithm. State its properties.
9. What is meant by image smoothing and image sharpening?
10. Give the two channel wavelet filter banks to decompose the input signal into frequency bands
.PART-B (5*16=80)
11. (a) 1. Define the following terms :
(i) Uniform noise.
(ii) Wiener – Khintchine theorem.
(iii) Power spectrum
(iv) Physical significance of spectral factorization.
2. Define Hilbert space and orthogonal projection . How does it help in estimation.
(Or)
(b) Consider a discrete random process X(n) = Acos(2?f0n+?)+?(n). where A and f0 are
constants , ? is a random variable with pdf f0 (?) = B 0=?=2?
= 0 elsewhere ?(n). is an independent white noise. Determine the autocorrelation and power
spectrum of this random process. And Assume a Gaussian random process with narrow band
spectrum zero mean and variance s2. Prove that the two quadrature components of it are
uncorrelated and possess equal variance.
12. (a) (i) Define cross correlation and cross spectrum. Relate the output cross spectrum in terms
of input cross spectrum.
(ii) Explain about smoothed spectral estimation.
(Or)
(b) Present the model based approach to power spectral estimation. Define AR,MA, and ARMA
models. And Illustrate the ARMA model for spectrum estimation.
13 (a) (i) Give the properties of linear estimators and the Cramer-Rao bound.
(ii) Briefly explain the estimation of a non stationary process by a Kalman filter.
(Or)
(b) (i) Describe the basics of forward linear prediction . Give the schematic of FIR filter and Lattice
filter for first order predictor.
(ii) Derive the recursive predictor coefficients for optimum lattice predictor by Levinson –Durbin
algorithm.
14(a) What do you understand by an adaptive filter? Discuss the minimum MSE criterion to
develop an adaptive FIR filter
(Or)
(b) Explain the adaptive channel equalization in detail.
15 (a) Define 2D DFT and state its separability and peridiocity properties. Explain its role in the
image smoothing and sharpening operations.
(Or)
(b) Explain the multisolutional analysis of wavelets . Explain the application of wavelets in signal
compression.
B.E / B.Tech, DEGREE EXAMINATION , NOV / DEC 2006
Fourth Semester (IT)
Computer Science and EngineeringVII
DIGITAL SIGNAL PROCESSING
( Common to B.E.(PartTime)
Fourth Semester Regulation 2005)
(Regulation 2004)
Time: 3 hours Maximum : 100 marks
Answer ALL Questions
PART A – (10 X 2 = 20 marks)
1. State Sampling Theorem?
2. Find the Poles of the system.
3. Find the DFT of the sequence x(n) ={1, 1, 0, 0 } DFT is obtained by FFT.
4. Calculate the number of multiplications needed in the calculation of 512 point radix
2FFT when compared to Direct DFT?
5. What are the properties that are maintained same in the transfer of analog filter into a
digital filter?
6. What is warping effect?

7.Draw the direct from realization of FIR system?

8.What are the describe features of a window function? Name the different types of
windowing function?
9. What is truncation?
10. Draw a sample/ hold circuit and explain its operations?
PART B – (5 X 16 = 80 marks)
11. (a) (i) For each of the following discrete time system, determine whether or not the
system is Linear Time, Variant, Causal and Stable?
11.(a) (ii) Determine the transfer function, magnitude & phase response, impulse
response for the system.
11. (b) (i) Find the Ztransform
of
1) x(n) = 2 n u (n2)
2) x(n) = n 2 u (n)
11.(b)(ii) Use convolution to find x(n), given
11.(b) (iii) Determine the cross correlation values of the sequence x1(n) ={ 1, 2, 3, 4 }
x2(n) = {4, 3, 2, 1}
12. (a) (i) Compute linear and circular convolution of the two sequence
x1(n) ={ 1, 2, 2, 2 } and x2(n) = {1, 2, 3, 4}
12.(a) (ii) Compute the FFT using DIT algorithm for the sequence x(n) = {1, 2, 3, 4, 4, 3,
2, 1 } and draw the corresponding flow diagram.
12.(b) (i)Prove that multiplication DFT’s of 2 sequence is equivalent to the DFT of the
circular convolution of the 2 sequence in time domain?
12.(b)(ii) Discuss in detail the use of FFT algorithm , in linear filtering?

13.(a) Find H(z) using impulse invariant technique for the analog system function.
13 (b) (i) Obtain the direct form II, Cascade form parallel form structures for the system?
13.(b) (ii) Design a butterworth filter using linear transformation that satisfies the
following constraint?
14(a) The desired response of a low pass filter is?
14.(b) Explain the Type I & Type 2 design of FIR filter using frequency sampling
Technique?
15.(a) The output of A/D converter is applied to a digital filter with
system function find the o/p noise power for the digital filter when the input
signal is quantized to 8Hz.
15.(b)(i) A digital system is characterized by the difference equation y(n)=0.95 y(n
1)+x(n). Determine the dead band the system when x(n) =0 and y(n) = 13
15.(b) (ii) With neat diagram explain the analysis and synthesis part of a recorder in
detail?

ANNA UNIVERSITY :: CHENNAI – 600 025

MODEL QUESTION PAPER
B.TECH. INFORMATION TECHNOLOGY
IF351 – DIGITAL SIGNAL PROCESSING

Time : 3 Hours Max. Marks : 100

Answer all Questions

PART – A (10 X 2 = 20 MARKS)

1. State and prove the convolution property of Z transform.

2. Check the system is linear or not y(n) = x(n)+ay(n-1)
3. Write equations for finding DFT and IDFT using Z transform.
4. Draw the radix 2 butterfly structure for DIF
5. Draw the implementation for the generalized for IIR filter using direct form II.
6. Explain how the addition and multiplication of (H1, H2) impulse responses implemented in filter
design
7. Write equations for Hanning and Blackman window.
8. Why frequency prewarping procedure is adopted in the design of IIR filter?
9. Write two advantages of musical sound processing and briefly explain.
10. Explain the effects due to upsampling.

PART - B (5 x 16 = 80 Marks)

11.i) The impulse response of a linear TI system is h(n) = {1, 0, 1, -1}. Find the response of the
system to the input signal x(n) = {1, 0, 2, 1}.

ii) Check whether the system y(n) = x(n) – x(n-1) is LTI and stable.

12.a) Develop and draw the 8 point radix-2 DIT FFT algorithm for DFT computation.

(OR)

12.b) Compute the DFT of the following sequence

x(n) = 0 0£ n £ 2
= 1 3£ n £ 6
= 0 n=7
Plot magnitude and phase spectra

13.a) Design a LPF with following specifications. Use Hamming window and at least 8 points.

(OR)

13.b)i) Obtain H(z) from H(s) when T = 1 sec.

ii) Design a digital BPF using w1 & w2 as cutoff frequencies

14.a)i) Perform the following using Floating Point arithmetic.
1.5 x 1.75 and 1.5 x 1.75

ii) Realize the following H(z) given by

using cascade and Parallel form with Direct form-I.

(OR)

14.b)i) What is meant by quantization error? Explain briefly.

ii) Realize the following filter using cascade technique with DF-I and DF-II.

15.a) Briefly explain

a. Interpolator
b. Decimator
c. Effects due to sampling rate conversion

(OR)

15.b)i) Write a note on Musical sound processing

ii) Explain how the data compression is achieved in speech signal and discuss a technique to
check the quality.
******