A Simple Nonorthogonal Approximate Joint Diagonalization Algorithm and Its Application for Blind Source Separation

被引:1
作者
Li, Changli [1 ]
Xu, Lizhong [1 ]
Xu, Shufang [1 ]
Zhang, Ying [2 ]
Tan, Yanchun [2 ]
机构
[1] Hohai Univ, Coll Comp & Informat Engn, Nanjing 211100, Jiangsu, Peoples R China
[2] Guangdong Ocean Univ, Sch Informat Engn, Zhanjiang 524088, Peoples R China
基金
中国国家自然科学基金;
关键词
Blind Source Separation (BSS); Joint Diagonalization; Approximate Joint Diagonalization; Diagonal Element;
D O I
10.1166/jctn.2015.3763
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Approximate joint diagonalization (AJD) of a set of matrices is an effective method for the blind source separation (BSS) problem. It is said that the separation performance of nonorthogonal AJD is better than that of orthogonal AJD, so this paper is devoted to designing a novel nonorthogonal AJD criterion. The proposed criterion does not impose any restriction on the diagonalizing matrix, but it can automatically exclude the degenerate solution during the learning iterations. It takes not only the off-diagonal elements but also the diagonal elements of diagonalized matrices into account, so it can achieve better diagonalization results. Simulation experiments are presented to demonstrate the validity of the proposed algorithm.
引用
收藏
页码:549 / 552
页数:4
相关论文
共 17 条
[1]  
Belouchrani A., 1997, IEEE T SIGNAL PROCES, V2, P45
[2]   Jacobi angles for simultaneous diagonalization [J].
Cardoso, JF ;
Souloumiac, A .
SIAM JOURNAL ON MATRIX ANALYSIS AND APPLICATIONS, 1996, 17 (01) :161-164
[3]  
Cichochi A., 2004, ADAPTIVE BLIND SIGNA
[4]  
Degerine S., 2007, IEEE T SIGNAL PROCES, V6, P55
[5]  
Fadaili E. M., 2007, IEEE T SIGNAL PROCES, V5, P55
[6]  
Li C., 2010, NEURAL COMPUT APPL, V2, P19
[7]  
Li C., 2014, J COMPUT THEOR NANOS, V2, P11
[8]  
Li X., 2007, IEEE T SIGNAL PROCES, V5, P55
[9]  
Narayanan M., 2012, QUANTUM MATTER, V1, P53, DOI DOI 10.1166/QM.2012.1005
[10]  
Ono T., 2012, QUANTUM MATTER, V1, P4, DOI DOI 10.1166/QM.2012.1002