Title :
Natural gradient for temporally dependent component analysis
Author :
Shi, Zhenwei ; Cheng, Dalong ; Tan, Xueyan ; Jiang, Zhiguo
Author_Institution :
Image Process. Center, Beihang Univ., Beijing, China
Abstract :
The temporally dependent component analysis (TDCA) method for blind source separation (BSS) is introduced. As a new principle, it is shown that maximizing the mapping of autocorrelation of source signals can be used to perform BSS. We use the natural gradient algorithm for TDCA and study the mathematical properties of TDCA. Simulations by square temporal autocorrelation sources verify the efficient implementation of the proposed method.
Keywords :
blind source separation; correlation methods; independent component analysis; blind source separation; natural gradient algorithm; source signals; square temporal autocorrelation sources; temporally dependent component analysis; Algorithm design and analysis; Blind source separation; Correlation; Independent component analysis; Signal processing algorithms; Stability analysis; Blind source separation (BSS); Independent component analysis (ICA); Linear autocorrelation; Nonlinear autocorrelation;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5583820
