Abstract
Fractional representation by polynomial matrices is a tool for describing linear dynamical systems, often providing a unique insight into the systems. It has turned out that the coefficient matrices in this representation can be identified without knowing the input data, under some statistic assumptions. This is an outcome by combining system theory with a recent progress in signal processing; i.e., a methodology generically called independent component analysis.
This article summarizes this technique of blind system identification. Restricted optimization for parameter estimation plays a key role. Also presented are some of its applications in various fields such as input distortion compensation, disturbance suppression, and time series prediction in financial engineering.
Keywords
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. John Wiley & Sons, Inc., Chichester (2001)
Cichocki, A., Amari, S.: Adaptive Blind Signal and Image Processing: Learning Algorithms and Applications. John Wiley & Sons, Inc., Chichester (2002)
Nitta, M., Sugimoto, K.: Blind identification of polynomial matrix fraction via independent component analysis. Int. J. of Robust and Nonlinear Control, Special Issue on Polynomial Methods 17, 732–751 (2007)
Nitta, M., Sugimoto, K.: Blind identification of MIMO discrete-time systems based on independent component analysis. IEICE Trans. J90-A, 27–34 (2007) (in Japanese); English translation is available in Electronics and Communications in Japan, Part II, vol. 90, pp. 17–25
Even, J., Sugimoto, K.: Estimation of MIMO system with nonlinear distortion at the input: an adaptive approach. In: Proc. SICE-ICASE International Joint Conference (SICE-ICCAS 2006), Busan, Korea, October 2006, pp. 3984–3989 (2006)
Sugimoto, K., et al.: Polynomial matrix approach to independent component analysis (Part II) application. In: Proc. 16th IFAC World Congr., FR-A22-TO/1, Prague, Czech Republic (July 2005); Journal version is available in Japanese as: Nitta, M., Sugimoto, K.: Disturbance rejection system based on independent component analysis. Trans. of the Society of Instrument and Control Engineers 42, 1313–1319 (2006); Nitta et al.: Blind identification of mechanical systems via independent componet analysis and detection of structure change. Trans. Inst. of Systems, Control and Information Engineers 19, 177–184 (2006)
Sugimoto, K., Kondo, H.: Multivariate time series prediction by blind signal deconvolution. In: Proc. ICROS-SICE International Joint Conference 2009, Fukuoka, Japan (August 2009)
Amari, S., Cichocki, A., Yang, H.: A new learning algorithm for blind signal separation. In: Advances in Neural Information Processing Systems, vol. 8, pp. 757–763. MIT Press, Cambridge (1996)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2010 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Sugimoto, K. (2010). Blind Identification of Polynomial Matrix Fraction with Applications. In: Willems, J.C., Hara, S., Ohta, Y., Fujioka, H. (eds) Perspectives in Mathematical System Theory, Control, and Signal Processing. Lecture Notes in Control and Information Sciences, vol 398. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-93918-4_34
Download citation
DOI: https://doi.org/10.1007/978-3-540-93918-4_34
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-93917-7
Online ISBN: 978-3-540-93918-4
eBook Packages: EngineeringEngineering (R0)