Natural gradient for temporally dependent component analysis
- Resource Type
- Conference
- Authors
- Shi, Zhenwei; Cheng, Dalong; Xueyan Tan; Jiang, Zhiguo
- Source
- 2010 Sixth International Conference on Natural Computation Natural Computation (ICNC), 2010 Sixth International Conference on. 2:972-975 Aug, 2010
- Subject
- Components, Circuits, Devices and Systems
Computing and Processing
Signal Processing and Analysis
Communication, Networking and Broadcast Technologies
General Topics for Engineers
Correlation
Algorithm design and analysis
Blind source separation
Signal processing algorithms
Independent component analysis
Stability analysis
Blind source separation (BSS)
Independent component analysis (ICA)
Linear autocorrelation
Nonlinear autocorrelation
- Language
- ISSN
- 2157-9555
2157-9563
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.