Abstract
This paper describes a generalized state space Blind Source Recovery (BSR) framework obtained by using the Kullback-Lieblar divergence as a performance functional and the application of optimization theory under the constraints of a feedforward state space structure. Update laws for both the non-linear and the linear dynamical systems have been derived for the domain of dynamic blind source recovery along both ordinary stochastic gradient and the Riemannian contra-variant gradient directions. The choice of the rich state space demixing network structure allows for the development of potent learning rules, capable of handling most filtering paradigms and convenient extension to non-linear models. Some particular filtering cases are subsequently derived from this general structure and are compared with material in the recent literature. Some of this reported work has also been implemented in dedicated hardware/software. An illustrative simulation example has been presented to demonstrate the online adaptation capabilities of the proposed algorithms.
Similar content being viewed by others
References
Amari, S.: Differential-geometrical methods in statistics; Lecture Notes in Statistics 1985. 28
Amari, S., Cichocki, A and Yang, H H.: A new learning algorithm for blind signal separation. In: Advances in Neural Information Processing Systems, MIT Press: Cambridge, MA. 1996, pp. 757–763
Amari, S., Douglas, S C., Cichocki, A and Yang, H H.: Multichannel Blind Deconvolution and Equalization using the Natural Gradient, In: Proceeding of IEEE Workshop on Signal Processing, Adv. in Wireless Communications, Paris, France 1997, pp. 101–104
Amari, S.: Natural Gradient Works Efficiently in Learning. In: Neural Computation, Vol. 10, 1998, pp. 251–276
Amari, S. and Nagaoka, H: Methods of Information Geometry, AMS and Oxford University Press, 1999
Araki, S, Makino, S, Mukai, R, Nishikawa, T and Saruwatari, H: Fundamental Limitation of Frequency Domain Blind Source Separation for Convolved Mixture of Speech; 3rd Int’l Symposium on ICA and BSS, ICA-2001. San Diego, CA. 2001, pp. 132–137
Cardoso, J F.: Entropic Contrasts for Source Separation. Geometry and Stability. In: Unsupervised Adaptive Filtering. Chapter 4, John Wiley & Sons, 1999
Cover, T M. and Thomas, J A.: Elements of Information Theory, Wiley Interscience Series, 1991
Erten, G and Salam, F M.: Voice Output Extraction by Signal Separation. In: Proceedings of the 1998 IEEE International Symposium on Circuits and Systems, 1998, pp. 5–8
Erten, G and Salam, F M.: Real Time Separation of Audio Signals Using Digital Signal Processors. In: Proceedings of the 40th Midwest Symposium on Circuits and Systems, 1998, pp. 1237–1240
Gharbi, A B. A. and Salam, F M.: Separation of Mixed Signals in Dynamic Environments: Formulation and some Implementation. In: Proceedings of 37th Midwest Symposium on Circuits and Systems, Lafayette, LA, 1994, pp. 587–590
A B. A. Gharbi F M. Salam (1995) ArticleTitleImplementation and Test Results of a Chip for the Separation of Mixed Signals IEEE Transactions on Circuits and Systems 42 IssueID11 748–751
Gharbi, A B. A. and Salam, F M.: Algorithms for Blind Signal Separation and Recovery in Static and Dynamic Environments. In: Proceedings of 1997 Int’l Symp. on Circuits and Systems, Hong Kong, 1997, pp. 713–716
Khalil, H K., Nonlinear Systems. 3rd ed. 2002: Prentice Hall
Lee, T W., Bell, A J. and Orglmeister, R: Blind Source Separation of Real-World Signals. In: Proceedings of IEEE Int’l Conference on Neural Networks ICNN-97, Houston, TX, 1997, pp. 2129–2135
Lee T W., Girolami M, Bell A J. and Sejnowski T J.: A Unifying Information-Theoretic Framework for Independent Component Analysis. In: Computers and Mathematics with Applications, (2000), 31(11): 1–21
Salam, F M.: An Adaptive Network for Blind Separation of Independent Signals. In: Proceedings of International Symposium on Circuits and Systems, Chicago, IL, 1993. pp. 431–434
Salam, F M. and Erten, G: Blind Signal Separation and Recovery in Dynamic Environments. In: Proceedings of 3rd IEEE Workshop on Nonlinear Signal and Image Processing, Mackinac Island, MI, 1997
Salam, F M. and Erten, G: The State Space Framework for Blind Dynamic Signal Extraction and Recovery. In: Proceedings of the IEEE International Symposium on Circuits and Systems, 1999, pp. 66–69
Salam, F M., Erten, G and Waheed, K: Blind Source Recovery: Algorithms for Static and Dynamic Environments. Proceedings of INNS-IEEE International Joint Conference on Neural Networks, 2001, pp. 902–907
Salam, F M. and Waheed, K: State-space feedforward and feedback structures for blind source recovery, In: Proceedings of 3rd International Conference on Independent Component Analysis and Blind Signal Separation, San Diego, CA, 2001, pp. 248–253
Waheed, K and Salam, F M.: Blind source recovery: some implementation and performance issues. Proceedings of the 44th IEEE Midwest Symposium on Circuits and Systems, Dayton, OH, 2001, pp. 694–697
Waheed, K and Salam, F M.: State-space blind source recovery for mixtures of multiple source distributions. In: Proceedings of IEEE Int’l Symposium on Circuits and Systems, Scottsdale, Arizona, 2002, pp. 197–200
Waheed, K and Salem, F M.: New hyperbolic source density models for blind source recovery score functions. In: IEEE International Symposium on Circuits and Systems, Bangkok, Thailand, 2003, Vol 3, pp. 32–35
Waheed, K and Salem, F M.: Blind source recovery: a state space formulation. Contemporary Issues in System Stability and Control with Applications, In: D. Liu and P. J. Antsaklis, (eds.), Birkhauser Boston, MA, 2003, pp. 171–196
Waheed, K and Salem, F M.: Blind source recovery: A framework in the state space; Journal of Machine Learning Research, Special Issue on Independent Component Analysis, the MIT Press, Cambridge, MA, USA, 4(Dec):1411–1446
Zhang, L and Cichocki, A: Blind separation/deconvolution of sources using canonical stable state-space models, In: Proceeding of the 1998 international Symposium on Nonlinear Theory and its Applications (NOLTA’98). Crans-Montana, Switzerland, 1998, pp. 927–930
Zhang, L and Cichocki, A: Blind deconvolution/equalization using state-space models. In: Proceedings of the 1998 IEEE Workshop on Neural Networks for Signal Processing (NNSP’98). Cambridge, UK, 1998, pp. 123–131
L Q. Zhang A Cichocki (2000) ArticleTitleBlind deconvolution of dynamical systems: A state space approach Journal of Signal Processing 4 IssueID2 113–130
Zhang, L Q., Cichocki, A and Amari, S.: Kalman filter and state-space approach to multichannel blind deconvolution. In: IEEE Workshop on Neural Networks for Signal Processing, Sydney, Australia, 2000, pp. 425–434
Author information
Authors and Affiliations
Corresponding authors
Rights and permissions
About this article
Cite this article
SALEM, F.M., WAHEED, K. & ERTEN, G. Blind Source Recovery in a State-Space Famework: Algorithms for Static and Dynamic Environments. Neural Process Lett 21, 153–173 (2005). https://doi.org/10.1007/s11063-004-5484-9
Received:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s11063-004-5484-9