Files in this item



application/pdfECE499-Sp2017-yadavalli.pdf (1MB)Restricted to U of Illinois
(no description provided)PDF


Title:Study of learning algorithms of adaptive filtering
Author(s):Yadavalli, Meghana
Contributor(s):Radhakrishnan, Chandrasekhar
Subject(s):adaptive filtering
FIR filters
LMS algorithm
particle swarm algorithm
Abstract:Adaptive filtering is a technique used to implement filtering in time-varying environments. The algorithms used to achieve this can be broadly classified as gradient descent methods and structured stochastic approaches. This work presents an overview of the different adaptive algorithms available to realize adaptive filters. Examples are given to illustrate that the least mean square (LMS) technique performs well in the context of adaptive FIR filters. Experiments to illustrate how the performance of LMS is affected by changing algorithm parameters and input conditioning are conducted in the context of FIR filters. Convergence issues when using the standard LMS based approach that can arise in IIR filters are also addressed. Finally, a structured stochastic approach called the Particle Swarm Algorithm is studied to show this algorithm has the potential to overcome some of the stability issues encountered when using gradient descent techniques on IIR filters.
Issue Date:2017-05
Date Available in IDEALS:2017-08-31

This item appears in the following Collection(s)

Item Statistics