Blind separation of noisy convolutive sources

In this paper, we present a new blind source separation method for noisy linear convolutive signal mixtures of independent components. The effectiveness of the proposed method is tested on synthetic data. Our technique has various advantages. First, it is based on the expectation-maximization (EM) algorithm for the separation of the components. Second, the proposed technique works in the spectral domain where, thanks to two simple approximations, the likelihood assumes a simple form which is easy to handle (low dimensional sufficient statistics) and to maximize (via the EM algorithm). Using this technique, we have obtained a good preliminary results being able to blindly separate noisy mixtures with two components and four different versions of mixing matrix.

[1]  Jingqing Luo,et al.  Using eigenvalue grads method to estimate the number of signal source , 2000, WCC 2000 - ICSP 2000. 2000 5th International Conference on Signal Processing Proceedings. 16th World Computer Congress 2000.

[2]  Hsien-Tsai Wu,et al.  Source number estimator using Gerschgorin disks , 1994, Proceedings of ICASSP '94. IEEE International Conference on Acoustics, Speech and Signal Processing.

[3]  Wai Lok Woo,et al.  Novel statistical approach to blind recovery of earth signal and source wavelet using independent component analysis , 2008 .

[4]  Lars Kai Hansen,et al.  Estimating the Number of Sources in a Noisy Convolutive Mixture Using BIC , 2004, ICA.

[5]  Heng-Ming Tai,et al.  Detection of the number of sources at low signal-to-noise ratio , 2007 .

[6]  Nicola Vanello,et al.  Comparative evaluation of decomposition algorithms based on frequency domain blind source separation of biomedical signals , 2005 .

[7]  Arti Khaparde,et al.  FastICA algorithm for the separation of mixed images , 2008 .

[8]  Mohammad Ali Tinati,et al.  An adaptive algorithm for speech source separation in overcomplete cases using wavelet packets , 2006 .

[9]  H. Vincent Poor,et al.  Estimation of the number of sources in unbalanced arrays via information theoretic criteria , 2005, IEEE Transactions on Signal Processing.

[10]  Hagit Messer,et al.  Order statistics approach for determining the number of sources using an array of sensors , 1999, IEEE Signal Processing Letters.

[11]  Terrence J. Sejnowski,et al.  Blind source separation of more sources than mixtures using overcomplete representations , 1999, IEEE Signal Processing Letters.

[12]  Thomas Kailath,et al.  Detection of signals by information theoretic criteria , 1985, IEEE Trans. Acoust. Speech Signal Process..

[13]  Bin He,et al.  Estimation of Number of Independent Brain Electric Sources From the Scalp EEGs , 2006, IEEE Transactions on Biomedical Engineering.

[14]  Markus Borschbach,et al.  Separation capability of overcomplete ICA approaches , 2007 .

[15]  Jun Wang,et al.  Blind source extraction from convolutive mixtures in ill-conditioned multi-input multi-output channels , 2004, IEEE Transactions on Circuits and Systems I: Regular Papers.

[16]  Damjan Zazula,et al.  Multichannel Blind Source Separation Using Convolution Kernel Compensation , 2007, IEEE Transactions on Signal Processing.