Design of Low-Cost FPGA Hardware for Real-time ICA-Based Blind Source Separation Algorithm

Blind source separation (BSS) of independent sources from their convolutive mixtures is a problem in many real-world multisensor applications. In this paper, we propose and implement an efficient FPGA hardware architecture for the realization of a real-time BSS. The architecture can be implemented using a low-cost FPGA (field programmable gate array). The architecture offers a good balance between hardware requirement (gate count and minimal clock speed) and separation performance. The FPGA design implements the modified Torkkola's BSS algorithm for audio signals based on ICA (independent component analysis) technique. Here, the separation is performed by implementing noncausal filters, instead of the typical causal filters, within the feedback network. This reduces the required length of the unmixing filters as well as provides better separation and faster convergence. Description of the hardware as well as discussion of some issues regarding the practical hardware realization are presented. Results of various FPGA simulations as well as real-time testing of the final hardware design in real environment are given.

[1]  M. Y. Siyal,et al.  Blind source separation of audio signals using improved ICA method , 2001, Proceedings of the 11th IEEE Signal Processing Workshop on Statistical Signal Processing (Cat. No.01TH8563).

[2]  Robert M. Gray,et al.  Entropy and Information , 1990 .

[3]  Wayne Luk,et al.  Parameterised floating-point arithmetic on FPGAs , 2001, 2001 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.01CH37221).

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

[7]  Ray Andraka,et al.  A survey of CORDIC algorithms for FPGA based computers , 1998, FPGA '98.

[8]  Te-Won Lee,et al.  Independent Component Analysis , 1998, Springer US.

[9]  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 .

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

[11]  C. Johnson,et al.  Theory and design of adaptive filters , 1987 .

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

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

[14]  Allan Kardec Barros,et al.  Heart instantaneous frequency (HIF): an alternative approach to extract heart rate variability , 2001, IEEE Transactions on Biomedical Engineering.

[15]  Allan Kardec Barros,et al.  Robust batch algorithm for sequential blind extraction of noisy biomedical signals , 1999, ISSPA '99. Proceedings of the Fifth International Symposium on Signal Processing and its Applications (IEEE Cat. No.99EX359).

[16]  Gert Cauwenberghs,et al.  Mixed-signal real-time adaptive blind source separation , 2004, 2004 IEEE International Symposium on Circuits and Systems (IEEE Cat. No.04CH37512).

[17]  Paris Smaragdis,et al.  Evaluation of blind signal separation methods , 1999 .

[18]  Farook Sattar,et al.  Low-cost design and implementation of an ICA-based blind source separation algorithm , 2002, 15th Annual IEEE International ASIC/SOC Conference.

[19]  Peiyu He,et al.  Exploiting Acoustic Similarity of Propagating Paths for Audio Signal Separation , 2003, EURASIP J. Adv. Signal Process..

[20]  Hiroshi Sawada,et al.  Blind source separation with different sensor spacing and filter length for each frequency range , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.