An Adaptive Subspace Algorithm For Blind Separation Of Independent Sources In Convolutive Mixture
Abstract
the advantage of the algorithm proposed in this coreespondance is that it reduces a convolutive mixture to an instantaneous mixture by using only second-order statistics (but more sensors than sources). Furthermore, the sources can be separated by using any algorithm applicable to an instantaneous mixture. Otherwise, to ensure the convergence of our algorithm, we assume some classical assumptions for blind separation of sources and some added subspace assumptions. Finally, the assumptions concerning the subspace model and their properties are emphasized in this correspondance. Index GTerms- Cholesky decomposition, convolutive mixture, ICA and Blind separation of sources, LMS, subspace methods, Sylvester matrix.