Abstract— In this paper, the canonical correlation approach for blind source separation is revisited. We proposed a novel criterion for source extraction, which is proved to be equivalent to the CCA criterion, whereas it has some advantages over the standard CCA criterion in developing fast learning algorithms. By optimizing the proposed criterion, we developed two learning algorithms for the extraction of specific signals. The first algorithm is based on the steepest descent technique, and the second one is a modified Newton algorithm. The experiment results demonstrate the effectiveness of the proposed learning algorithms, and show that the modified Newton algorithm converges much faster than the other extraction method.
Index Terms— Blind source separation, canonical correlation analysis, steepest descent, Newton iteration.
Wei-Tao Zhang, Xiao-Guang Yuan, and Shun-Tian Lou are with the School of Electronic Engineering, Xidian University, Xi’an 710071, China (e-mail: zhwt-work@foxmail.com, xgyuan@xidian.edu.cn, loushuntian@gmail.com).
[PDF]
Cite: Wei-Tao Zhang, Xiao-Guang Yuan, and Shun-Tian Lou, " A CCA Criterion Based Adaptive Algorithm for Blind Extraction of Specific Signal," International Journal of Information and Electronics Engineering vol. 4, no. 5, pp. 370-374, 2014.
