An independent component analysis algorithm using signal noise ratio

被引：0

作者：

Wan, Jun ^{[1
]}

Zhang, Xiao-Hui ^{[1
]}

Rao, Jiong-Hui ^{[1
]}

Hu, Qing-Ping ^{[1
]}

机构：

[1] Department of Weapon Engineering, Naval University of Engineering, Wuhan 430033, China

来源：

Kongzhi yu Juece/Control and Decision | 2012年 / 27卷 / 09期

关键词：

Algorithms - Blind source separation - Computer simulation;

D O I：

暂无

中图分类号：

学科分类号：

摘要：

An independent component analysis (ICA) algorithm based on the combination of negentropy and signal noise ratio (SNR) is presented to solve the deficiency of traditional ICA method after the introduction of principle and algorithm of ICA. The main formulas in the algorithm are elaborated and the idiographic steps of algorithm are given. Then the computer simulation is used to testify the performance of this algorithm. Both the traditional fast ICA algorithm and the presented ICA algorithm are applied to separate the mixed signal data. Experiment results show that the proposed method has better performance on separating signals than the traditional fast ICA algorithm based on negentropy, and can estimate the source signals from the mixed signals more exactly.

引用

页码：1376 / 1380