Scribd
Upload
6 views3 pages

SSP Jan 2022

This document is a question paper for the M.E./M.Tech. degree examinations in January 2022, focusing on Communication Systems and Statistical Signal Processing. It includes various questions divided into three parts, covering topics such as stationary random processes, power spectral density, estimation methods, and adaptive filters. The paper consists of both short answer questions and detailed problems requiring derivations and explanations.

Uploaded by

039 Leena Gifty S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
6 views3 pages

SSP Jan 2022

This document is a question paper for the M.E./M.Tech. degree examinations in January 2022, focusing on Communication Systems and Statistical Signal Processing. It includes various questions divided into three parts, covering topics such as stationary random processes, power spectral density, estimation methods, and adaptive filters. The paper consists of both short answer questions and detailed problems requiring derivations and explanations.

Uploaded by

039 Leena Gifty S
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 3

Question Paper Code : 12118

M.E./M.Tech. DEGREE EXAMINATIONS,JANUARY 2022.

First Semester

Communication Systems
DS 4152 -- STATISTICALSIGNAL PROCESSING
(Common to : M.E. Communiation and Networking/M.E. Digital Signal Processing)

(Regulations 2021)
Time : Three hours Maximum : 100 marks
Answer ALL questions.

PART A— (10 x 2 = 20 marks)

1. What is stationary random process and how does it differ from the wide sense
stationary process?

2. What is power spectral density of a random process and what information does
it provide about the process?

3. What is meant by consistent estimate?

4. What is the difference between ML estimate and MAP estimate?

5. What is peridogram estimator?

6. What are the limitations Of n0n-parametric methods for egtimating power


spectrum?

7. What is meant by forward and backward linear prediction?

8. Bring out the difference between Ralman filter and Wiener filter.

9. What is an adaptive filter and mention its applications?

10. A zero-mean white-noise Of variance is the input to the


steepest-descent algorithm• What is the condiuon for the
in mean convergence of the
steepest descent algorithm square?
PART B — (5 x 13 = 65 marks)

11. (a) A causal LSI system, which is described by the difference equation

.y(n) —1),is driven by a zero-mean WSS


process x(n)with autocorrelation rx Determine the power
spectral density and autocorrelation of the output sequence (13)
Or

(b) Show that the necessary and sufficient condition for a WSS random
is
process x(n) to be ergodic in mean (13)

lim I (I-Ll

Where the auto-covarianceof x(n)

12. (a) Let and be two independent unbiased estimators of parameter


Ohaving variances and c; respectively. Consider an estimator of
the form 0=aÖ1+bÖ2. Determinethe values of a and b so that the
estimate 0 is unbiasedand has minimum variance. (13)
Or
(b) Consider an N independentobservations x i, i=l, 2,...N, of a Gaussian
random variablewith mean g and unknown variance . Find the
maximumlikelihoodestimate of the variance . Compute the bias and
variance of the estimate.
(13)

13. (a) Showthat the periodogram


power
estimator is not a consistent estimator for
spectrum estimation.
(13)

Or
(b) With necessary expressions,
spectrum, explain Welch method for estimating power
(13)
(a) Derive the Wiener-Hopf
hence derive the minimumequations for optimum causal IIR filter and
mean square error (MMSE), (13)
Or
(b) Derive I.evinsion•Durbin
predictor coefficients algorithm for recursively updating forward
and hence develop lattice structure for linear
(13)

2 12118
15. (a) Derive normalized
LMS algorithm for an Nth order FIR adaptive filter.
(13)

Or
(b) Draw a block diagram of
adaptive noise canceller (ANC)and show that
minimizing mean square error
its output signal power. of the ANC is equivalent to maximizing
(13)

PART C— (1 x 15 = 15 marks)

16. (a) Starting from the basic principleand with relevant equations, explain
how an estimation of signal is done using discrete of Kalman filter. (15)

Or
(b) The autocorrelation values for an autoregressive (AR) process x(n) of
order-2 for lag values k=0, 1, and 2 are given as
rx rx and rx
The AR process is generated by the white noise input v(n) with mean
zero and variance
(i) Determine the AR(2)modelparameters and the variance of the
input white noise. (6)
(ii) Determine the numerical value of rÄ(3) (5)
(iii) Determine the power spectrum of the process. (4)

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy