Author

Xu Li

Publication Date

1989

Document Type

Dissertation/Thesis

First Advisor

Kuo, Sen M. (Sen-Maw)

Degree Name

M.S. (Master of Science)

Department

Department of Electrical Engineering

LCSH

Adaptive filters||Adaptive signal processing

Abstract

In an effort to explore a better approach to improve the performance of frequency domain adaptive filter, a new transform, the Fast Hartley Transform (FHT), is proposed for the transform domain adaptive filter. This new filter was implemented on the AT&T DSP 16 digital signal processor (DSP) for real-time applications. The analysis was then taken to evaluate the use of the FHT as the transforms replacing the FFT in the frequency-domain adaptive filter (FDAF). This innovation has shown a distinct improvement by measuring the performance of the Hartley transform domain adaptive filter (HDAF) and comparing against its Fast Fourier Transform counterpart as well as its corresponding time-domain adaptive filters. This thesis presents a comprehensive description on the Hartley transform domain adaptive signal processing under the following topics of interest: the HDAF structures, properties, algorithms, and analysis; the FHT and its applications in adaptive signal filtering; the implementation of the HDAF on digital signal processor; and the performance comparisons between filter structures and transforms. There are two principal advantages to the Hartley-domain implementations of adaptive filters. First, the amount of computation required to process a fixed set of data can be greatly reduced compared with time-domain approaches. Second, the convergence properties of the adaptive process can be improved over simple gradient descent method. Other properties are also discussed. The C language and DSP assembly programs for computer simulations and the simulation results are developed in this thesis.

Comments

Includes bibliographical references (pages 61-62)

Extent

viii, 114 pages

Language

eng

Publisher

Northern Illinois University

Rights Statement

In Copyright

Rights Statement 2

NIU theses are protected by copyright. They may be viewed from Huskie Commons for any purpose, but reproduction or distribution in any format is prohibited without the written permission of the authors.

Media Type

Text

Share

COinS