Third-order cumulants based method for continuous-time Errors-In-Variables system identification by fractional models

Author

Chetoui, M. ; Malti, R. ; Thomassin, M. ; Aoun, M. ; Najar, S. ; Abdelkrim, M.N.

Author_Institution

Lab. de l´´Integration du Materiau au Syst., Univ. de Bordeaux, Talence, France

fYear

2011

fDate

22-25 March 2011

Firstpage

1

Lastpage

6

Abstract

This paper deals with continuous-time system identification using fractional models in a noisy input/output context. The third-order cumulants based least squares method (tocls) is extended here to fractional models. The derivatives of the third-order cumulants are computed using a new fractional state variable filter. A numerical example is used to demonstrate the performance of the proposed method called ftocls (fractional third-order cumulants based least squares). The effect of the signal-to-noise ratio and the hyperparameter is studied.

Keywords

continuous time systems; differential equations; identification; least squares approximations; continuous-time system; errors-in-variables system; fractional models; fractional state variable filter; hyperparameter effect; least squares method; noisy input context; noisy output context; signal-to-noise ratio effect; system identification; third-order cumulants based method; Equations; Least squares methods; Mathematical model; Noise measurement; Numerical models; Signal to noise ratio; Simulation; Errors-In-Variables; Fractional SVF; Fractional derivatives; High-order statistics; System identification; Third-order cumulants;

fLanguage

English

Publisher

ieee

Conference_Titel

Systems, Signals and Devices (SSD), 2011 8th International Multi-Conference on

Conference_Location

Sousse

Print_ISBN

978-1-4577-0413-0

Type

conf

DOI

10.1109/SSD.2011.5767441

Filename

5767441