源信号数目未知的自然梯度盲信号分离算法

›› 2015, Vol. 28 ›› Issue (3): 79-.

源信号数目未知的自然梯度盲信号分离算法

邵莲莲

(西安电子科技大学数学与统计学院,陕西西安 710071)

出版日期:2015-03-15 发布日期:2015-03-12
作者简介:邵莲莲(1987—),女,硕士研究生。研究方向:统计学习算法,盲信号与信息处理中的数学方法。E-mail:shaolianlian2012@163.com

Natural Gradient Algorithm for Blind Source Separation with Unknown Source Number

SHAO Lianlian

(School of Mathematics and Statistics,Xidian University,Xi'an 710071,China)

Online:2015-03-15 Published:2015-03-12

摘要/Abstract

摘要：

总结了源信号数目未知的盲信号分离自然梯度算法,得到自然梯度算法发散的原因,分离矩阵的各行沿混合矩阵转置的零空间方向无效的冗余移动。借助投影自然梯度算法,从理论上证明,冗余分量的范数随迭代次数的增加呈指数分布。

关键词: 盲信号分离, 自然梯度算法, 冗余分量, 指数分布

Abstract:

A summary of the natural gradient algorithm for blind source separation with an unknown number of sources is presented and the conclusion is drawn that it is the redundant movement of separating matrix in the zero space of the transpose of mixing matrix that causes the divergence.By the projected natural gradient algorithm,we theoretically prove that with the increase in the number of iterations,the norm of the redundant subjects to an exponential distribution.

Key words: blind source separation;natural gradient algorithm;the redundant;exponential distribution

中图分类号:

TN911.7

邵莲莲. 源信号数目未知的自然梯度盲信号分离算法[J]. , 2015, 28(3): 79-.

SHAO Lianlian. Natural Gradient Algorithm for Blind Source Separation with Unknown Source Number[J]. , 2015, 28(3): 79-.

参考文献

［1］Hyvarinen A,Karhunen J,Oja E.Independent component analysis［M］.New York:Wiley,2001.

［2］Ji Min Y,Xiao Long Z,Xian Da Z,Adaptive blind separation with an unknown number of sources［J］.Neural Computation,2004,16(8):1641-1660.

［3］Cichocki A,Karhunen J,Kasprzak W,et al.Neural networks for blind separation with unknown number of sources［J］.Neuro Computing,1999,24(2):55-93.

［4］冶继民,张贤达,朱孝龙.信源数目未知和动态变化时的盲信号分离［J］.中国科学:E辑,2005,35(12):1-12.

［5］冶继民,金海红,楼顺天,等.未知源信号数目投影自然梯度盲信号分离算法［J］.西安电子科技大学学报:自然科学版,2006,33(2):190-194.

［6］张贤达,保铮.盲信号分离［J］.电子学报,2001,29(12):1766-1771.

［7］Amari S.Natural gradient works efficiently in learning［J］.Neural Computation,1998,10(2):251-276.

［8］Amari S.Natural gradient for over-and under-complete bases in ICA［J］.Neural Computation,1999,11(8):1875-1883.

［9］Xiao Long Z,Xian Da Z,Ji Min Y,Natural gradient-based recursive least squares algorithm for blind source separation［J］.Sci.China Ser.F,2004,47(1):55-65.

［10］Amari S.Neural learning in structured parameter space:Neural Riemannian gradient［M］.Cambridge,MA:MIT Press,1997.

［11］Hyvarinen A.Fast and robust fixed-point algorithms for independent component analysis［J］.IEEE Transactions on Neural Networks,1999,10(3):626-634.

［12］Cichocki A,Sabala I,Choi S,et al.Self-adaptive independent component analysis for sub-Gaussian and super-Gaussian mixtures with unknown number of source signals and additive noise［C］.International Sympily on Nonlinear Theory and Applications,1997.

［13］金海红,冶继民.源信号个数未知的超定盲信号分离的半参数统计算法［J］.西安电子科技大学学报:自然科学版,2003,30(6):844-848.