Instructor
Srikrishna Bhashyam
Office: ESB2 405
Phone: 2257 4439
Pre-requisites
Probability Foundations for Electrical Engineers (EE5110 or EE3110).
Course Description
This is a graduate-level course on adaptive filters. The design and performance of adaptive filters are discussed. Two classes of algorithms -- stochastic gradient algorithms and least squares algorithms -- to adapt the coefficients of a linear filter are discussed in detail. The topics covered are:
1) Review of Estimation Theory
--- Minimum Mean Squared Error (MMSE) estimation
--- Linear MMSE estimation
--- Sequential linear MMSE estimation
--- Kalman filter
2) Stochastic Gradient Algorithms
--- Least Mean Squares (LMS) Algorithm
--- Mean-square performance
--- Transient performance
3) Least Squares Algorithms
--- Recursive Least Squares (RLS) algorithm
--- Kalman filtering and RLS algorithm
4) Other topics from:
--- Array Algorithms
--- Lattice Filters
--- Robust Filters
--- Other performance criterion (other than MMSE and LS)
References
[1] A. H. Sayed, Adaptive Filters, John Wiley & Sons, NJ, ISBN 978-0-470-25388-5, 2008. Video Lectures here
[2] S. Haykin, Adaptive Filter Theory, Fourth Edition, Pearson Education LPE, 2007.
[3] Alexander D. Poularikas, Zayed M. Ramadan, Adaptive filtering primer with MATLAB, CRC Press, 2006.
[4] B. Widrow and S.D. Stearns, Adaptive Signal Processing, Prentice Hall, Englewood Cliffs, NJ, 1985.
Lecture Notes
2018 Lecture Notes
Evaluation
As per Institute Academic Calendar
Quiz 1 (20%) -- Aug 30, 2023
Assignment (15%)
Project (20%)
Final (45%) -- Nov 20, 2023