Improvements in Adaptive IIR Filtering: Theory and Application

Characteristics of the mean-square error surface in adaptive digital filters determine how well a gradient algorithm performs within a given filter structure, i.e., if the surface has steep slopes and contains local minima, a gradient algorithm will have difficulty reaching the global minimum. It is shown that although Stearns' conjecture holds strictly for first- and second-order filters, it is not true in general, and that an additional restriction introduced by Soderstrom is needed for unimodality of the error surface. The adverse effect of overparameterization which can have serious practical implications is shown through an example. Also, it is shown that for certain insufficient order filters, a nonminimum phase characteristic is sufficient for multimodality of the error surface when the unknown system is driven by white noise. A convenient method for finding the stationary points is introduced.

Name thesis:degree_name

Ph.D.

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Electrical Engineering

Grantor

University of Illinois at Urbana-Champaign

Year dc:date

2014

Author dc:creator

• Nayeri, Majid

• Engineering, Electronics and Electrical

Identifier

(UMI)AAI8823211

OAI identifier oai:identifier

oai:www.ideals.illinois.edu:2142/69400