Huan Zhao

Publication Date

1990

Document Type

Dissertation/Thesis

First Advisor

Kuo, Sen M. (Sen-Maw)

Degree Name

M.S. (Master of Science)

Legacy Department

Department of Electrical Engineering

LCSH

Adaptive signal processing; Electronic noise

Abstract

Acoustic echoes existing in teleconferencing, video-conference, and hands-free telephony systems and acoustic echo cancellation methods utilizing adaptive signal processing techniques are investigated in this work. The primary goals of the work have been developing optimized adaptive algorithms and filter structures for acoustic echo cancellation as well as bridging the gap between theoretical algorithm researches and practical real-time implementations. It is a comprehensive study of applying adaptive signal processing algorithms and structures to the cancellation of acoustic echoes. Starting with a brief introduction to the basic structures and algorithms digital adaptive filtering systems and an analysis of the performance and convergence behavior of adaptive network echo cancellers, the special characteristics of acoustic echoes and difficulties of cancelling them with conventional echo cancellation algorithms are then analyzed. Based on the analysis, algorithms and structures for acoustic echo cancellation are studied and a transform domain echo canceler structure is developed. Besides the algorithms development, practical issues involved in an acoustic echo canceler such as double talk treatment and speech detection are investigated based on the realworld situation. Real-time implementation considerations, using both fixed-point and floating- point digital signal processors (DSPs), are studied and many useful guidelines are presented in the thesis. A compact hands-free telephone design accomplished at the end of the work as an application of the theories resulted from the work.

Comments

Includes bibliographical references (pages [196]-204)

Extent

xv, 244 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

Download

Share

COinS