An Algorithm Based on Nonlinear PCA for Blind Separation of Convolutive Mixtures

MA Li-yan; LI Hong-wei

您当前的位置：

首页 >

文章列表页 >

An Algorithm Based on Nonlinear PCA for Blind Separation of Convolutive Mixtures

更新时间：2025-07-16

- An Algorithm Based on Nonlinear PCA for Blind Separation of Convolutive Mixtures
- Acta Electronica Sinica Vol. 36, Issue 5, Pages: 1009-1012(2008)
- 作者机构：
  
  中国地质大学数理学院,湖北,武汉,430074
- 作者简介：
- 基金信息：
- DOI：
  CLC： TN911.7
- Published：2008
- 稿件说明：
移动端阅览
MA Li-yan, LI Hong-wei. An Algorithm Based on Nonlinear PCA for Blind Separation of Convolutive Mixtures[J]. Acta Electronica Sinica, 2008, 36(5): 1009-1012.
DOI：

MA Li-yan, LI Hong-wei. An Algorithm Based on Nonlinear PCA for Blind Separation of Convolutive Mixtures[J]. Acta Electronica Sinica, 2008, 36(5): 1009-1012. DOI：

摘要

本文针对卷积混合的盲源分离

首先提出了一种基于非线性主成分分析(PCA)的准则

然后推导出了一种递推最小二乘(RLS)算法.该算法为时域盲源分离算法

分为两个阶段

第一阶段为混合信号的白化

第二阶段为基于白化信号的RLS迭代.新算法不存在频域算法在每个频点的幅度与次序不确定性的问题

具有较快的收敛速度.文中最后通过仿真试验验证了本文算法的有效性.

Abstract

This paper proposes a nonlinear principle component analysis (PCA)cost function for blind separation of convolutive mixtures.Then a new recursive least-squares (RLS) algorithm is developed in time domain.This algorithm has two stages:one is whitening

the other is RLS iteration.Also

the algorithm does not suffer from the problems of frequency domain based algorithms on arbitrary scaling and permutations in different frequency components.Simulations show that this algorithm can successfully separate mixing signals and has fast convergence.

关键词

Keywords

references

Views

989

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

A New Method Based on the Full Utilizations of Concerning Variablesc Structures for Blind Source Separation of Convolutive Mixtures

Blind Source Separation with Space-Time Analytical CMA in CDMA Smart Antenna Systems

Related Author

XU Xian-feng

FENG Da-zheng

LIU Yan-hong

WEI Xue-ye

JI Song-po

FENG Ai-gang

YIN Qin-ye

DENG Ke

Related Institution

National Key Laboratory for Radar Signal Processing,Xidian University

School of Electronics and Information Engineering, Beijing Jiaotong University

China Telecom Xinjiang Telecom Co.Ltd.

School of Electronics and Information EngineeringBeijing Jiaotong UniversityBeijing 100044China

China Telecom Xinjiang Telecom Co.Ltd.UrumqiXinjiang 830026China

⁰