Abstract
The paper presents the application of the successive linearisation to the neural network implementation of the Volterra filter. Applying the signal flow graph approach, the new learning rules for adaptation of weights of the multilayer network structure obtained are given. The multilayer structure presented is applied to signal processing, including identification of the parameters of the plant, noise cancelling and signal prediction. The results of the simulation of the filter in these three basic application modes are presented and discussed.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Widrow B, Steams S. Adaptive Signal Processing, Prentice Hall, New York, 1985
Haykin S. Adaptive Filter Theory, Prentice Hall, New York, 1991
Rauf F, Ahmed HM. Successive linearization: a new approach to non-linear adaptive filtering. Workshop ‘Higher Order Statistics’, Chamrousse, 1991, 293–296
Kuh T, Powers EJ. Second order Volterra filtering and its application to nonlinear system identification. IEEE Trans Acoustics, Speech and Signal Processing 1985;33:1445–1455
Mathews VJ. Adaptive polynomial filters. IEEE SP Magazine 1991:10–26
Hertz J, Krough A, Palmer R. Introduction to the Theory of Neural Computation, Addison Wesley, Amsterdam, 1991
Osowski S. Signal flow graphs and neural networks. Biological Cybernetics 1994;70:387–395
Stark J. Recursive prediction of chaotic time series. J Nonlinear Science 1993:197–223
Osowski S, Vu Quang T. Multilayer neural network structure as Volterra filter. IEEE Proc ISCAS, London 1994:253–256
TUTSIM-Simulation for Design and Analysis, Reference Manual, 1988
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Osowski, S. Multilayer volterra filter and its applications. Neural Comput & Applic 4, 228–236 (1996). https://doi.org/10.1007/BF01413821
Issue Date:
DOI: https://doi.org/10.1007/BF01413821