The Maximum Likelihood Approach to Complex ICA

We derive the form of the best non-linear functions for performing independent component analysis (ICA) by maximum likelihood estimation. We show that both the form of nonlinearity and the relative gradient update equations for likelihood maximization naturally generalize to the complex case, and that they coincide with the real case. We discuss several special cases for the score function as well as adaptive scores.

[1]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .

[2]  Paris Smaragdis,et al.  Blind separation of convolved mixtures in the frequency domain , 1998, Neurocomputing.

[3]  Terrence J. Sejnowski,et al.  Complex Independent Component Analysis of Frequency-Domain Electroencephalographic Data , 2003, Neural Networks.

[4]  Vince D. Calhoun,et al.  Complex Infomax: Convergence and Approximation of Infomax with Complex Nonlinearities , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[5]  N. N. Vakhania,et al.  Random Vectors with Values in Complex Hilbert Spaces , 1997 .

[6]  Visa Koivunen,et al.  Complex ICA for circular and non-circular sources , 2005, 2005 13th European Signal Processing Conference.

[7]  Vince D. Calhoun,et al.  Independent component analysis by complex nonlinearities , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[8]  Aapo Hyvärinen,et al.  A Fast Fixed-Point Algorithm for Independent Component Analysis of Complex Valued Signals , 2000, Int. J. Neural Syst..

[9]  V. Koivunen,et al.  Ieee Workshop on Machine Learning for Signal Processing Complex-valued Ica Using Second , 2022 .

[10]  Jean-François Cardoso,et al.  Equivariant adaptive source separation , 1996, IEEE Trans. Signal Process..

[11]  Philippe Garat,et al.  Blind separation of mixture of independent sources through a quasi-maximum likelihood approach , 1997, IEEE Trans. Signal Process..