论文信息 - Sparse Source Separation with Unknown Source Number

Sparse Source Separation with Unknown Source Number

Sparse Blind Source Separation (BSS) problems have recently received some attention. And some of them have been proposed for the unknown number of sources. However, they only consider the overdetermined case (i.e. with more sources than sensors). In the practical BSS, there are not prior assumptions on the number of sources. In this paper, we use cluster and Principal Component Analysis (PCA) to estimate the number of the sources and the separation matrix, and then make the estimation of sources. Experiments with speech signals demonstrate the validity of the proposed method.

Hongwei Li | Rui Qi | Yujie Zhang

[1] P. Comon,et al. Blind Identification of Overcomplete MixturEs of sources (BIOME) , 2004 .

[2] Christian Jutten,et al. Estimating the mixing matrix in Sparse Component Analysis (SCA) based on partial k-dimensional subspace clustering , 2008, Neurocomputing.

[3] Terrence J. Sejnowski,et al. Learning Overcomplete Representations , 2000, Neural Computation.

[4] Xie Sheng. Underdetermined Blind Source Separation Algorithm Based on Normal Vector of Hyperplane , 2008 .

[5] S. Haykin,et al. Adaptive Filter Theory , 1986 .

[6] Richard M. Everson,et al. Independent Component Analysis: Principles and Practice , 2001 .

[7] Juha Karhunen,et al. Neural networks for blind separation with unknown number of sources , 1999, Neurocomputing.

[8] Christian Jutten,et al. Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[9] Christian Jutten,et al. Sparse ICA via cluster-wise PCA , 2006, Neurocomputing.

[10] Barak A. Pearlmutter,et al. Blind source separation by sparse decomposition , 2000, SPIE Defense + Commercial Sensing.

[11] Barak A. Pearlmutter,et al. Independent Component Analysis: Blind source separation by sparse decomposition in a signal dictionary , 2001 .