A Characteristic Function Based Contrast Function for Blind Extraction of Statistically Independent Signals

Muhammad TUFAIL  Masahide ABE  Masayuki KAWAMATA  

Publication
IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer Sciences   Vol.E89-A   No.8   pp.2149-2157
Publication Date: 2006/08/01
Online ISSN: 1745-1337
DOI: 10.1093/ietfec/e89-a.8.2149
Print ISSN: 0916-8508
Type of Manuscript: Special Section PAPER (Special Section on Papers Selected from the 20th Symposium on Signal Processing)
Category: 
Keyword: 
independent component analysis,  characteristic function,  non-Gaussianity measure,  contrast function,  

Full Text: PDF(362KB)
>>Buy this Article


Summary: 
In this paper, we propose to employ a characteristic function based non-Gaussianity measure as a one-unit contrast function for independent component analysis. This non-Gaussianity measure is a weighted distance between the characteristic function of a random variable and a Gaussian characteristic function at some adequately chosen sample points. Independent component analysis of an observed random vector is performed by optimizing the above mentioned contrast function (for different units) using a fixed-point algorithm. Moreover, in order to obtain a better separation performance, we employ a mechanism to choose appropriate sample points from an initially selected sample vector. Finally, some computer simulations are presented to demonstrate the validity and effectiveness of the proposed method.