首页 | 本学科首页   官方微博 | 高级检索  
     

基于粒子群算法的盲源分离算法
引用本文:张文爱,刘俊豪. 基于粒子群算法的盲源分离算法[J]. 太原理工大学学报, 2006, 37(2): 169-172
作者姓名:张文爱  刘俊豪
作者单位:太原理工大学,信息工程学院,山西,太原,030024
摘    要:简要地介绍了粒子群算法(PSO)及其改进算法和盲源分离算法(BSS),改进的粒子群算法具有并行性、易实现等优点。将改进的粒子群算法与盲源分离算法相结合,提出了基于粒子群算法的盲源分离算法。该算法以混合信号的峰度为目标函数,采用独立分量分析的方法,用改进的粒子群算法代替常规的最陡梯度下降法,对瞬时混合的信号进行盲分离,解决了梯度算法收敛速度慢的问题。实验仿真表明:该算法具有收敛速度快、分离效果好等特点。

关 键 词:盲源分离  粒子群算法  群集智能
文章编号:1007-9432(2006)02-0169-04
收稿时间:2005-07-10
修稿时间:2005-07-10

Blind Source Separation Based on Particle Swarm Optimization
ZHANG Wen-ai,LIU Jun-hao. Blind Source Separation Based on Particle Swarm Optimization[J]. Journal of Taiyuan University of Technology, 2006, 37(2): 169-172
Authors:ZHANG Wen-ai  LIU Jun-hao
Affiliation:College of Information Engineering of TUT , Taiyuan 030024 ,China
Abstract:In this paper,blind source separation,particle swarm optimization and modified particle swarm optimization are introduced and a new algorithm of blind source separation based on particle swarm optimization is proposed.Using kurtosis of the mixed signal to the target function of blind source separation,replacing steepest gradient descent method by modified particle swarm optimization,it succeeds to separate the instantaneous mixed signal by the method of independent component analysis.It solves the problem of low searching speed in steepest gradient descent method.The computer simulation results showed its character of fast convergence speed and stable performance,etc.
Keywords:blind source separation  particle swarm optimization  swarm intelligence
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号