130 likes | 150 Views
This course covers the fundamentals of adaptive signal processing including linear algebra, filters, signal estimation, and detection. Topics include Wiener filters, LMS and RLS algorithms, support vector machines, and pattern recognition. References included.
E N D
EE491DSpecial Topics in CommunicationsAdaptive Signal Processing Spring 2005 Prof. Anthony Kuh POST 205E Dept. of Elec. Eng. University of Hawaii Phone: (808)-956-7527, Fax: (808)-956-3427 Email: kuh@spectra.eng.hawaii.edu
Preliminaries • Class Meeting Time: MWF 11:30-12:20 • Office Hours: MWF 10-11 (or by appointment) • Prerequisites: • Probability and Random Variables: EE342 or equivalent • Digital Signal Processing: EE 415 can be taken concurrently • Programming: Matlab or C experience
Objectives and Grading Topics: Adaptive signal processing. Objectives: Understand basic concepts, applications. Design project chosen from text or literature synthesizing basic ideas. Grading: • Homework: 25% • Exam:25% • Final project: 50% (oral presentation and written report)
Overview of Course Material • Background Material • Linear Algebra • Vector and Matrix operations • Eigenvalues and Eigenvectors • Probability and Random Variables • Gaussian Random vectors, Stationary processes, 2nd order processes • Discrete time filters • Matlab
Overview Continued • Optimum Filtering • Estimation and Detection • Mean Squared Error Criterion, Energy surface • Wiener Filter • Steepest Descent • Algorithm • Convergence and Step Size
Overview Continued • Least Mean Square (LMS) Algorithm • Algorithm • Convergence and step size • Applications • Variations • Least Square Algorithms • Algorithm • Properties • Applications
Overview Continued • Recursive Least Square (RLS) Algorithms • Algorithm • Convergence and behavior • Applications • Variants
Overview Continued • Kernel Methods • Kernel transformation • Optimization • Least squares support vector machine • Support vector regression
Overview Continued • Pattern recognition • Linear threshold unit: Perceptron Learning Algorithm • Optimum Margin Classifiers • Support Vector Machine
Overview Continued • Other Topics • Component analysis: Principal Component Analysis (PCA), Kernel PCA, Independent Component Analysis, Blind Source Separation • Multilayer feedforward networks: Error backpropagation algorithm • Linear prediction and Kalman Filtering
References • S. Haykin. Adaptive Filter Theory 4th Ed. Prentice Hall, Englewood Cliffs, NJ, 2001. • B. Widrow and S. Stearns. Adaptive Signal Processing. Prentice Hall, Englewood Cliffs, NJ, 1985. • S. Haykin. Neural Networks, A comprehensive foundation, 2nd Ed. Prentice Hall, Englewood Cliffs,NJ, 1998.
What is Signal Processing? ``The theory and application of filtering, coding, transmitting, estimating, detecting, analyzing, recognizing, synthesizing, recording, and reproducing signals by digital or analog devices or techniques. “Signal" includes audio, video, speech, image, communication, geophysical, sonar, radar, medical, musical, and other signals’’ IEEE Signal Processing Society
Why ``Adaptive’’ Signal Processing? • System or channel characteristics are unknown. • System or channel characteristics are time varying.