Overcomplete BSS with a Geometric Algorithm Export

@misc{citeulike:1632342, abstract = {We present an independent component analysis (ICA) algorithm based on geometric considerations [10] [11] to decompose a linear mixture of more sources than sensor signals. Bofill and Zibulevsky [2] recently proposed a two-step approach for the separation: first learn the mixing matrix, then recover the sources using a maximum-likelihood approach. We present an e\#cient method for the matrix-recovery step mimicking the standard geometric algorithm thus generalizing Bofill and Zibulevsky's...}, author = {Theis, F. J. and Lang, E. W. and Westenhuber, T. and Puntonet, C. G.}, citeulike-article-id = {1632342}, citeulike-linkout-0 = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.16.7552}, keywords = {bss, ica, overcomplete}, posted-at = {2007-09-07 19:25:40}, priority = {2}, title = {Overcomplete BSS with a Geometric Algorithm}, url = {http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.16.7552} }

TY - GEN ID - citeulike:1632342 L3 - citeulike-article-id:1632342 N2 - We present an independent component analysis (ICA) algorithm based on geometric considerations [10] [11] to decompose a linear mixture of more sources than sensor signals. Bofill and Zibulevsky [2] recently proposed a two-step approach for the separation: first learn the mixing matrix, then recover the sources using a maximum-likelihood approach. We present an e#cient method for the matrix-recovery step mimicking the standard geometric algorithm thus generalizing Bofill and Zibulevsky's... TI - Overcomplete BSS with a Geometric Algorithm KW - bss KW - ica KW - overcomplete AU - Theis, FJ AU - Lang, EW AU - Westenhuber, T AU - Puntonet, CG UR - http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.16.7552 ER -