EE 504

Adaptive Signal Processing

Spring 2004

 

Instructor: Cagatay Candan

Email: ccandan -at- metu.edu.tr

Office: C-105

 

Course Outline:  We will be discussing the methods of filtering stationary and non-stationary signals in this course. The filtered signal is used at many applications including echo-cancellation, system modeling, channel equalization etc. We will examine Wiener, least-square filtering methods for the stationary/deterministic signals; and LMS/RLS adaptive filtering methods and Kalman filters for the non-stationary signals.  Familiarity with the concepts of random processes and linear algebra is expected.

 

·        Introduction

o      Review of Random Processes

o      Mean Square Estimation Techniques, (Linear MSE estimation, optimal estimation)

o      Filtering the Random Processes

o      Moving Average (MA), Auto-regressive (AR) and ARMA processes

 

·        Wiener Filtering (Solving Wiener-Hopf Equations)

 

·        Review of Iterative Methods in Linear Equation System Solving

 

o     Method of Steepest Descent

 

·        Iterative Methods for Solving Wiener-Hopf Equation System 

o     LMS Method and its variants

o     FIR , IIR LMS Filters

o     RLS Filters

o     Kalman Filtering

 

·        Applications

o     Channel Equalization, Linear Predictive Coding, Echo-Cancellation

 

References:

1. Simon Haykin, Adaptive Filter Theory, Prentice Hall, 1996.
2. Monson H. Hayes, Statistical Digital Signal Processing and Modelling, John Wiley & Sons, 1996.
3. Athanasios Papoulis, Probability, Random Variables, and Stochastic Processes, Mc-Graw Hill, 1991.
   
   

Grading: Two midterms and final,  homeworks with Matlab assignments .