Performance analysis of Volterra kernel estimators with Gaussian inputs

Author

Redfern, Arthur J. ; Zhou, G. Tong

Author_Institution

Sch. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA

fYear

1997

fDate

21-23 Jul 1997

Firstpage

162

Lastpage

165

Abstract

The focus of this paper is on Volterra nonlinear system identification from input-output data. When the system is linear-quadratic and the input is Gaussian, closed-form expressions for the kernels were derived by Tick (1961) based on input-output cross-cumulants. However, there have been no known variance expressions for the kernel estimates. In this paper, we analyze the performance of the first- and second-order kernel estimates when the input is zero-mean white Gaussian, and the additive noise has unknown color and distribution. Closed-form variance expressions are presented and verified by simulations

Keywords

Gaussian processes; Volterra series; estimation theory; higher order statistics; identification; noise; nonlinear systems; signal processing; Gaussian inputs; Volterra kernel estimators; Volterra nonlinear system identification; additive noise; closed-form expressions; closed-form variance expressions; color; first-order kernel estimates; input-output cross-cumulants; input-output data; linear-quadratic system; performance analysis; second-order kernel estimates; zero-mean white Gaussian; Additive noise; Analysis of variance; Closed-form solution; Color; Colored noise; Kernel; Noise measurement; Nonlinear systems; Performance analysis; Statistical analysis;

fLanguage

English

Publisher

ieee

Conference_Titel

Higher-Order Statistics, 1997., Proceedings of the IEEE Signal Processing Workshop on

Conference_Location

Banff, Alta.

Print_ISBN

0-8186-8005-9

Type

conf

DOI

10.1109/HOST.1997.613508

Filename

613508