MAP-Based Underdetermined Blind Source Separation of Convolutive Mixtures by Hierarchical Clustering and ℓ1-Norm Minimization. (5th September 2006)

• Record Type: Journal Article
• Authors: Winter Winter, Stefan Stefan; Kellermann Kellermann, Walter Walter; Sawada Sawada, Hiroshi Hiroshi; Makino Makino, Shoji Shoji
• Other Names: Ehlers Ehlers Frank Frank Academic Editor.
• Is Part Of: EURASIP journal on advances in signal processing. Volume 2007(2007)
• Publication Date: 2006-09-05
• Subjects: Signal processing -- Periodicals; Traitement du signal; Signal processing; Periodicals; 621.3822
• DOI: 10.1155/2007/24717
• Languages: English
• ISSNs: 1687-6172

Abstract

Abstract : We address the problem of underdetermined BSS. While most previous approaches are designed for instantaneous mixtures, we propose a time-frequency-domain algorithm for convolutive mixtures. We adopt a two-step method based on a general maximum a posteriori (MAP) approach. In the first step, we estimate the mixing matrix based on hierarchical clustering, assuming that the source signals are sufficiently sparse. The algorithm works directly on the complex-valued data in the time-frequency domain and shows better convergence than algorithms based on self-organizing maps. The assumption of Laplacian priors for the source signals in the second step leads to an algorithm for estimating the source signals. It involves theℓ 1 -norm minimization of complex numbers because of the use of the time-frequency-domain approach. We compare a combinatorial approach initially designed for real numbers with a second-order cone programming (SOCP) approach designed for complex numbers. We found that although the former approach is not theoretically justified for complex numbers, its results are comparable to, or even better than, the SOCP solution. The advantage is a lower computational cost for problems with low input/output dimensions.