Two-Step Sparse Component Analysis for Underdetermined Blind Source Separation
LI Ying-sheng · Audio Engineering · 2010
A sparse decomposition approach used in underdetermined BSS(Blind Source Separation)is presented.First,sparse representation(factorization) of a data matrix is discussed.Next,BSS is discussed based on sparse factorization approach.The blind separation technique includes two steps:estimating a mixing matrix and estimating sources.If the sources are sufficiently sparse,blind separation can be carried out directly in the time domain.Otherwise,blind separation can be implemented in time-frequency domain after applying wavelet packet transformation preprocessing to the observed mixtures.Theoretical considerations are supported by simulation results showing theoretical analysis is correct.