Blind signal separation in teleconferencing using ICA mixture model

被引：10

作者：

Bae, UM

Lee, TW

Lee, SY

机构：

[1] Korea Adv Inst Sci & Technol, Dept Elect Engn, Yusong Gu, Taejon 305701, South Korea

[2] Korea Adv Inst Sci & Technol, Brain Sci Res Ctr, Yusong Gu, Taejon 305701, South Korea

[3] Salk Inst Biol Studies, Computat Neurobiol Lab, La Jolla, CA 92037 USA

[4] Univ Calif San Diego, Inst Neural Computat, San Diego, CA 92103 USA

来源：

ELECTRONICS LETTERS | 2000年 / 36卷 / 07期

关键词：

Mathematical models - Matrix algebra - Signal processing;

D O I：

10.1049/el:20000459

中图分类号：

TM [电工技术]; TN [电子技术、通信技术];

学科分类号：

0808 ; 0809 ;

摘要：

An algorithm for separating several sources with fewer sensors in a nonstationary environment is presented. The nonstationary environment is. modelled by tracking unmixing matrix classes and the optimal class number is determined using a modified split and merge algorithm. Its usefulness is demonstrated by considering a teleconferencing problem.

引用

页码：680 / 682

页数：3