An Efficient Pairwise Kurtosis Optimization Algorithm for Independent Component Analysis

In the framework of Independent Component Analysis (ICA), kurtosis has been used widely in designing source separation algorithms. In fact, the sum of absolute kurtosis values of all the output components is an effective objective function for separating arbitrary sources. In this paper, we propose an efficient ICA algorithm via a modified Jacobi optimization procedure on the kurtosis-sum objective function. The optimal rotation angle for any pair of the output components can be solved directly. It is demonstrated by numerical simulation experiments that our proposed algorithm can be even more computationally efficient than the FastICA algorithm under the same separation performance.

[1]  Nathalie Delfosse,et al.  Adaptive blind separation of independent sources: A deflation approach , 1995, Signal Process..

[2]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[3]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[4]  Ying Tan,et al.  Advances in Neural Networks - ISNN 2008, 5th International Symposium on Neural Networks, ISNN 2008, Beijing, China, September 24-28, 2008, Proceedings, Part I , 2008, ISNN.

[5]  Jean-Francois Cardoso,et al.  Blind signal separation: statistical principles , 1998, Proc. IEEE.

[6]  Asoke K. Nandi,et al.  Blind separation of independent sources for virtually any source probability density function , 1999, IEEE Trans. Signal Process..

[7]  Eric Moreau,et al.  High order contrasts for self-adaptive source separation criteria for complex source separation , 1996 .

[8]  Jean-Franois Cardoso.  High-Order Contrasts for Independent Component Analysis , 1999, Neural Computation.

[9]  Asoke K. Nandi,et al.  Combined estimation scheme for blind source separation with arbitrary source PDFs , 2001 .

[10]  Fei Ge,et al.  Analysis of the Kurtosis-Sum Objective Function for ICA , 2008, ISNN.

[11]  E. Oja,et al.  Independent Component Analysis , 2001 .

[12]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[13]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..