An information theoretic approach to statistical linearization

Author

Chernyshov, K.R.

Author_Institution

Inst. of Control Sci., Moscow, Russia

fYear

2001

fDate

4-7 Sept. 2001

Firstpage

3654

Lastpage

3658

Abstract

The paper presents a procedure to derive a linear input/output model which is a statistical equivalent of a nonlinear dynamic stochastic system driven by a Gaussian white-noise input process. The key issue of such a procedure is using a statistical linearization criterion which is the condition of coincidence of the mutual information of the input and output processes of the system and the mutual information of the input and output processes of the system model. The approach provides obtaining explicit relationships which determine the weight coefficients of the linearized model.

Keywords

Gaussian noise; information theory; linear systems; linearisation techniques; nonlinear dynamical systems; statistical analysis; stochastic systems; white noise; Gaussian white-noise input process; information theoretic approach; linear input/output model; linearized model; nonlinear dynamic stochastic system; output process; statistical equivalent; statistical linearization criterion; weight coefficient; Decision support systems; Europe; Three-dimensional displays; Zinc; Measures of dependence; Mutual entropy; Shannon mutual information; Statistical linearization; System identification;

fLanguage

English

Publisher

ieee

Conference_Titel

Control Conference (ECC), 2001 European

Conference_Location

Porto

Print_ISBN

978-3-9524173-6-2

Type

conf

Filename

7076501