On the identification of additive nonlinear systems

A certain nonlinear system of an adaptive form is identified. Each branch of the system consists of a nonlinear gain followed by a dynamic linear subsystem. Algorithms for recovering characteristics of the system are proposed. The linear subsystem is identified by the correlation technique, whereas the nonlinear part is recovered with the help of orthogonal function techniques for estimation of the regression function. This is carried out by solving a system of integral equations relating to nonlinear characteristics and regression functions. The applied orthogonal functions are related to a diagonal expansion of the input joint probability density function. The conditions for consistency are established and the rate of convergence is evaluated.<<ETX>>

[1]  R. Tibshirani,et al.  Generalized Additive Models , 1991 .

[2]  W. Greblicki,et al.  Nonparametric Estimation of a Class of Nonlinear Time Series Models , 1991 .

[3]  Wκodzimierz Greblicki,et al.  Hermite Series Estimates of a Probability Density and its Derivatives , 1984 .

[4]  J. F. Barrett,et al.  An expansion for some second-order probability distributions and its application to noise problems , 1955, IRE Trans. Inf. Theory.

[5]  Miroslaw Pawlak,et al.  Nonparametric identification of a cascade nonlinear time series system , 1991, Signal Process..

[6]  Adam Krzyzak,et al.  On estimation of a class of nonlinear systems by the kernel regression estimate , 1990, IEEE Trans. Inf. Theory.

[7]  W. Greblicki,et al.  Fourier and Hermite series estimates of regression functions , 1985 .

[8]  W. Greblicki,et al.  Nonparametric identification of two-channel nonlinear systems , 1986, 1986 25th IEEE Conference on Decision and Control.

[9]  W. Greblicki,et al.  Hammerstein system identification by non-parametric regression estimation , 1987 .

[10]  Miroslaw Pawlak,et al.  Nonparametric identification of Hammerstein systems , 1989, IEEE Trans. Inf. Theory.

[11]  John B. Thomas,et al.  A characterization of Markov sequences , 1975 .

[12]  A. Krzyżak.  Identification of discrete Hammerstein systems by the Fourier series regression estimate , 1989 .

[13]  H. O. Lancaster.  The Structure of Bivariate Distributions , 1958 .

[14]  G. L. Wise,et al.  On a completeness property of series expansions of bivariate densities , 1975 .

[15]  W. Greblicki,et al.  Identification of discrete Hammerstein systems using kernel regression estimates , 1986 .

[16]  W. Greblicki.  Non-parametric orthogonal series identification of Hammerstein systems , 1989 .

[17]  W. Greblicki,et al.  Recursive nonparametric identification of Hammerstein systems , 1989 .

[18]  C. J. Stone,et al.  Additive Regression and Other Nonparametric Models , 1985 .

[19]  M. Pawlak.  On the series expansion approach to the identification of Hammerstein systems , 1991 .