An efficient algorithm for extracting single independent component

Author

Feng Da-Zhang ; Zheng, Bao ; Hai-Qin, Zhang ; Xian-Da, Zhang

Author_Institution

Lab. of Radar Signal Process., Xidian Univ., Xi´´an, China

Volume

1

fYear

2000

fDate

6/22/1905 12:00:00 AM

Firstpage

421

Abstract

We present a new off-line algorithm for extracting the independent components one by one from simultaneous mixer corrupted by noise, which relies only on a new criterion and an efficient search algorithm to achieve the extraction of the single independent component. The algorithm is an efficient, off-line, update approach, and can find the global optimal solution of the cost function defined in this paper. The advantage of this algorithm over existing methods is its ability to separate sources in spatially colored noise

Keywords

array signal processing; eigenvalues and eigenfunctions; least squares approximations; random noise; statistical analysis; cost function; efficient search algorithm; eigenvectors; global optimal solution; least squares approach; off-line algorithm; sensor array processing; single independent component extraction; source separation; spatially colored noise; Colored noise; Cost function; Covariance matrix; Crosstalk; Independent component analysis; Iterative algorithms; Least squares approximation; Least squares methods; Signal processing; Signal processing algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing Proceedings, 2000. WCCC-ICSP 2000. 5th International Conference on

Print_ISBN

0-7803-5747-7

Type

conf

DOI

10.1109/ICOSP.2000.894522

Filename

894522