EC8091 - ADVANCED DIGITAL SIGNAL PROCESSING (Syllabus) 2017-regulation Anna University

EC8091 - ADVANCED DIGITAL SIGNAL PROCESSING (Syllabus) 2017-regulation Anna University

EC8091	ADVANCED DIGITAL SIGNAL PROCESSING	LPTC

3003

OBJECTIVES:

• To learn and understand the concepts of stationary and non-stationary random signals and analysis & characterization of discrete-time random processes
• To enunciate the significance of estimation of power spectral density of random processes
• To introduce the principles of optimum filters such as Wiener and Kalman filters
• To introduce the principles of adaptive filters and their applications to communication engineering
• To introduce the concepts of multi-resolution analysis

UNIT I	DISCRETE-TIME RANDOM PROCESSES	9

Random variables - ensemble averages a review, random processes - ensemble averages, autocorrelation and autocovariance matrices, ergodic random process, white noise, filtering random processes, spectral factorization, special types of random processes - AR, MA, ARMA

UNIT II	SPECTRUM ESTIMATION	10

Bias and consistency, Non-parametric methods - Periodogram, modified-Periodogram - performance analysis. Bartlett's method, Welch's method, Blackman-Tukey method. Performance comparison. Parametric methods - autoregressive (AR) spectrum estimation - autocorrelation method, Prony's method, solution using Levinson Durbin recursion.

UNIT III	OPTIMUM FILTERS	9

Wiener filters - FIR Wiener filter - discrete Wiener Hopf equation, Applications - filtering, linear prediction. IIR Wiener filter - causal and non-causal filters. Recursive estimators - discrete Kalman filter.

UNIT IV	ADAPTIVE FILTERS	9

Principles and properties of adaptive filters - FIR adaptive filters. Adaptive algorithms - steepest descent algorithm, the LMS algorithm - convergence. Applications of adaptive filtering - noise cancellation, channel equalization.

UNIT V	MULTIRESOLUTION ANALYSIS	9

Short-time Fourier transform - Heisenberg uncertainty principle. Principles of multi-resolution analysis - sub-band coding, the continuous and discrete wavelet transform - properties. Applications of wavelet transform - noise reduction, image compression.

TOTAL: 45 PERIODS

OUTCOMES: At the end of the course, the student should be able to:

• Articulate and apply the concepts of special random processes in practical applications
• Choose appropriate spectrum estimation techniques for a given random process
• Apply optimum filters appropriately for a given communication application
• Apply appropriate adaptive algorithm for processing non-stationary signals
• Apply and analyse wavelet transforms for signal and image processing based applications

TEXT BOOKS:

1. Monson H. Hayes, "Statistical digital signal processing and modeling", John Wiley and Sons Inc. New York, Indian reprint 2008. (UNIT I-IV)
2. P. P. Vaidyanathan, "Multirate systems and filter banks", Prentice Hall Inc. 1993 (UNIT V)

REFERENCES:

1. John G. Proakis & Dimitris G.Manolakis, ―Digital Signal Processing – Principles, Algorithms & Applications‖, Fourth Edition, Pearson Education / Prentice Hall, 2007.
2. Sophoncles J. Orfanidis, "Optimum signal processing", McGraw Hill, 2000