Selective attention system using new active noise controller

被引：4

作者：

Sohn, JI

Lee, M

机构：

[1] Kyungpook Natl Univ, Sensor Technol Res Ctr, Puk Gu, Taegu 702701, South Korea

[2] Kyungpook Natl Univ, Dept Sensor Engn, Puk Gu, Taegu 702701, South Korea

来源：

NEUROCOMPUTING | 2000年 / 31卷 / 1-4期

关键词：

active noise control; blind source separation; selective attention system;

D O I：

10.1016/S0925-2312(99)00171-X

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

A new active noise control system is proposed that can selectively cancel a particular noise signal from a mixture. Blind source separation of the desired signal from the other noise signals is performed by a dynamic recurrent neutral network that is used as a preprocessor of this active noise control system. The proposed active noise control system then adaptively generates an anti-noise signal to specifically remove only the separated noise signal. Computer experimental results show that the proposed scheme can be effective in the construction of a selective attention system. (C) 2000 Elsevier Science B.V. All rights reserved.

引用

页码：197 / 204

页数：8

共 10 条

[1] AN INFORMATION MAXIMIZATION APPROACH TO BLIND SEPARATION AND BLIND DECONVOLUTION [J].