Overcomplete Independent Component Analysis via Linearly Constrained Minimum Variance Spatial Filtering
- 95 Downloads
Independent Component Analysis (ICA) designed for complete bases is used in a variety of applications with great success, despite the often questionable assumption of having N sensors and M sources with N≥M. In this article, we assume a source model with more sources than sensors (M>N), only L<N of which are assumed to have a non-Gaussian distribution. We argue that this is a realistic source model for a variety of applications, and prove that for ICA algorithms designed for complete bases (i.e., algorithms assuming N=M) based on mutual information the mixture coefficients of the L non-Gaussian sources can be reconstructed in spite of the overcomplete mixture model. Further, it is shown that the reconstructed temporal activity of non-Gaussian sources is arbitrarily mixed with Gaussian sources. To obtain estimates of the temporal activity of the non-Gaussian sources, we use the correctly reconstructed mixture coefficients in conjunction with linearly constrained minimum variance spatial filtering. This results in estimates of the non-Gaussian sources minimizing the variance of the interference of other sources. The approach is applied to the denoising of Event Related Fields recorded by MEG, and it is shown that it performs superiorly to ordinary ICA.
Keywordsindependent component analysis blind source separation overcomplete underdetermined EEG MEG denoising event related fields event related potentials beamforming
Unable to display preview. Download preview PDF.
- 1.R. Boscolo, H. Pan, and V. P. Roychowdhury, “Beyond Comon’s Identifiability Theorem for Independent Component Analysis,” Artificial Neural Networks—ICANN 2002 Lecture Notes in Computer Science, vol. 2415, 2002, pp. 1119–1124.Google Scholar
- 2.A. W. Bowman and A. Azzalini, Applied Smoothing Techniques for Data Analysis, Oxford University Press, 1997.Google Scholar
- 4.J. F. Cardoso and A. Souloumiac, “Blind Beamforming for Non–Gaussian Signals,” IEE Proc. F, vol. 140, 1993, pp. 362–370.Google Scholar
- 8.J. Eriksson and V. Koivunen, “Identifiability and Separability of Linear ICA Models Revisited,” in 4th International Symposium on Independent Component Analysis and Blind Signal Separation (ICA 2003), 2003, pp. 23–27.Google Scholar