Adaptive and Statistical Signal Processing

Course Description:

The field of adaptive filters and systems constitutes an important part of statistical signal processing. An adaptive system alters or adjusts its defining parameters in such a way that it improves performance through contact with the environment. Adaptive filters are currently applied in such diverse fields as communications, control, radar, seismology, and biomedical electronics. This course will cover the theory and applications of adaptive linear systems. Topics include Wiener filters, least squares, steepest descent, LMS, RLS, Newton's method, FIR and IIR adaptive structures, and Kalman filters. Applications covered include noise canceling, signal enhancement, adaptive control, adaptive beam-forming, system identification, and adaptive equalization.This course should be of interest to electrical and computer engineers specializing in signal processing and the information sciences. This course should also be taken as background for additional classes offered in artificial neural networks, connectionist models, and machine learning.

Course handouts:

• Student Information Sheet
• Information Sheet
• Homework:
• Homework Solutions:
• Lecture Summaries:
  • Introduction
  • Probability and Random Processes
  • Filtering Random Processes
  • Wiener Filtering
  • Wiener Filter - Prediction Example
  • Power Spectrum
  • Least Squares
  • Search Methods
  • LMS
  • System ID
  • Flitered-X and Adjoint LMS
  • Prediction
  • RLS
  • Noise Cancellation
  • Eigenvalue spread
  • Block LMS
  • Adaptive Equalizers
  • Kalman Filtering (talk) (intro paper)
• Example Student Projects:
  • Titles