一种基于AR模型的非线性盲源提取方法及其应用

山东大学学报(工学版) ›› 2010, Vol. 40 ›› Issue (5): 17-23.

一种基于AR模型的非线性盲源提取方法及其应用

蔡英1, 王刚2*

1. 四川托普信息技术职业学院, 四川成都 611743; 2. 电子科技大学电子工程学院, 四川成都 611731

收稿日期:2010-04-23 出版日期:2010-10-16 发布日期:2010-04-23
通讯作者: 王刚(1977-)，男，辽宁葫芦岛人,讲师，博士，主要研究方向为盲信号处理. E-mail:E-mail:wanggang-hld@hotmail.com
作者简介:蔡英(1975-)，女，四川成都人，讲师，主要研究方向未嵌入式系统. E-mail: caitangying@163.com
基金资助:
国家自然科学基金资助项目(30900318)

An AR parameters-based source selection method in general nonlinear blind source extraction

CAI Ying 1, WANG Gang 2*

1. Sichuan TOP Vocational Institute of Information Technology, Chengdu 611743, China;
2. School of Electronic Engineering,University of Electronic Science and Technology of China, Chengdu 610054, China

Received:2010-04-23 Online:2010-10-16 Published:2010-04-23

摘要/Abstract

摘要：

针对非线性盲源分离中非线性问题转化为线性问题,提出了一种基于AR模型的新方法。该方法在已知源信号的AR模型前提下,不但能够处理源信号的分离问题,还能够提取特定源信号,而后者是原来方法不具备的。从语音信号的非线性混合中提取源信号的仿真实验证实了该算法的有效性。

关键词: 非线性盲源分离, 非线性盲源提取, 均方预测误差, 协预测误差, AR 模型

Abstract:

In the nonlinear blind source separation (BSS) case,an AR parameters method was introduced as a new selection procedure. Compared with previous methods, the proposed algorithm not only can do the separation, but also can extract any desired signal with the corresponding AR parameter. It can deal with nonlinear blind source extraction (BSE) at the cost of more prior information and its performance is demonstrated on nonlinearly mixed speech data.

Key words: nonlinear blind source extraction, nonlinear blind source extraction, mean square prediction error (MSPE), mean cross prediction error (MCPE), AR parameter

蔡英1, 王刚2*. 一种基于AR模型的非线性盲源提取方法及其应用[J]. 山东大学学报(工学版), 2010, 40(5): 17-23.

CAI Ying 1, WANG Gang 2*. An AR parameters-based source selection method in general nonlinear blind source extraction[J]. JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE), 2010, 40(5): 17-23.

参考文献

多维度评价

Viewed

Full text

434

HTML			PDF

Just accepted	Online first	Issue	Just accepted	Online first	Issue
0	0	0	0	0	434

From	Others	local

Times	4	430
Rate	1%	99%

Abstract

680

Just accepted	Online first	Issue

0	0	680

	From	Others

	Times	680
	Rate	100%

Cited

Web of Science	Crossref	ScienceDirect	Search for Citations in Google Scholar >>


This page requires you have already subscribed to WoS.

Shared

Discussed