Blind source separation of audio signals using improved ICA method

Blind source separation (BSS) of independent sources from their convolutive mixtures is a problem in many real-world multi-sensor applications. In this paper, we propose an improved BSS method for audio signals based on ICA (independent component analysis) technique. It is performed by implementing non-causal filters instead of causal filters within the feedback network of the ICA based BSS method. It reduces the required length of the unmixing filters considerably as well as providing better results and faster convergence compared to the case with the conventional causal filters. The proposed method has been simulated and compared for real world audio signals.

[1]  R. Gray.  Entropy and Information Theory , 1990, Springer New York.

[2]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[3]  Reinhold Orglmeister,et al.  Blind source separation of real world signals , 1997, Proceedings of International Conference on Neural Networks (ICNN'97).

[4]  Kari Torkkola,et al.  Blind separation of convolved sources based on information maximization , 1996, Neural Networks for Signal Processing VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop.

[5]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[6]  Harold H. Szu,et al.  Independent component analysis approach to resolve the multi-source limitation of the nutating rising-sun reticle based optical trackers , 2000 .