The Transactions of the Korean Institute of Electrical Engineers D (대한전기학회논문지:시스템및제어부문D)

The Korean Institute of Electrical Engineers (대한전기학회)
권호 표지

Utilization of the Filtered Weighted Least Squares Algorithm For the Adaptive Identification of Time-Varying Nonlinear Systems

적응 FWLS 알고리즘을 응용한 시변 비선형 시스템 식별

  • 안규영 (한양대 전자통신전파 공학과) ;
  • 이인환 (한양대 전자전기컴퓨터 공학부) ;
  • 남상원 (한양대 전자전기컴퓨터 공학부)
  • Published : 2004.12.01
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, the problem of adaptively identifying time-varying nonlinear systems is considered. For that purpose, the discrete time-varying Volterra series is employed as a system model, and the filtered weighted least squares (FWLS) algorithm, developed for adaptive identification of linear time-varying systems, is utilized for the adaptive identification of time-varying quadratic Volterra systems. To demonstrate the performance of the proposed approach, some simulation results are provided. Note that the FWLS algorithm, decomposing the conventional weighted basis function (WBF) algorithm into a cascade of two (i.e., estimation and filtering) procedures, leads to fast parameter tracking with low computational burden, and the proposed approach can be easily extended to the adaptive identification of time-varying higher-order Volterra systems.

Keywords

References

  1. M. Niedzwieckim, Identification of Time-varying Processes, Wiley, New York, 2000
  2. R. Haber and L. Keviczky, Nonlinear System Identification-Input-Output Modeling Approach, Kluwer Academic Publishers, 1999
  3. S. Haykin, Adaptive Filter Theory, 4th Ed., Prentice-Hall, Inc., Englewood Cliffs, NJ, 2002
  4. M. Niedzwiecki and T. Klaput, 'Fast recursive basis function estimators for identification of time-varying processes,' IEEE Trans. Signal Processing, vol. 50, pp. 1925-1934, Aug. 2002 https://doi.org/10.1109/TSP.2002.800390
  5. M. Schetzen, The Volterra and Wiener Theories of Nonlinear Systems, John Wiley and Sons, Inc., New York, 1980
  6. F. Yuan and A. Opal, 'Distortion analysis of periodically switched nonlinear circuits using time-varying Volterra series,' IEEE Trans. Circuits Syst, vol. 48, pp. 726-738, Jun. 2001 https://doi.org/10.1109/81.928155
  7. W. Yu, S. Sen, and B.S. Leung, 'Distortion analysis of MOS track-and-Hold sampling mixers using time-varying Volterra series,' IEEE Trans. Circuits Syst.-II: Analog and Digital Signal Proc, vol. 46, no. 2, pp. 101-113, Feb. 1999 https://doi.org/10.1109/82.752910
  8. M. Iatrou, T.W. Berger, and V.Z. Marmarelis, 'Modeling of nonlinear nonstationary dynamic systems with a novel class of artificial neural networks,' IEEE Trans. Neural Networks, vol. 10, no. 2, pp. 327-339, Mar. 1999 https://doi.org/10.1109/72.750563
  9. V.J. Mathews and G.L. Sicuranza, Polynomial Signal Processing, John Wiley & Sons, Inc., 2000
  10. V.J. Mathews, 'Adaptive polynomial filters,' IEEE Signal Proc. Mag., vol. 8, no. 3, pp. 10-26, Jul. 1991 https://doi.org/10.1109/79.127998
  11. M. Niedzwiecki, 'First-order tracking properties of weighted least squares estimators,' IEEE Trans. Automat. Contr., vol. 33, pp. 94-96, Jan. 1988 https://doi.org/10.1109/9.365