Computer and Information Technology, International Conference on (2004)
Sept. 14, 2004 to Sept. 16, 2004
Shuxue Ding , University of Aizu
Jie Huang , University of Aizu
Daming Wei , University of Aizu
Sadao Omata , Nihon University
In this paper, we investigate some issues related to real-time processing for independent component analysis (ICA), based on gradient learning with simultaneous perturbation stochastic approximation (SPSA). Real-time ICA processing is especially necessary for an application in dynamic mixing environment, since a batch type of ICA processing can work well only in a static or stationary mixing environment. Although there are many choices for an ICA object function to which SPSA can be applied, in this paper, we choose a diagonality of the non-linear correlation matrix as our object function. Theories and implementations of the algorithm are described. Results of computer simulation are also presented to demonstrate the effectiveness.
S. Ding, D. Wei, S. Omata and J. Huang, "Signal Extensions in Independent Component Analysis and Its Application for Real-Time Processing," Computer and Information Technology, International Conference on(CIT), Wuhan, China, 2004, pp. 839-844.