Direct Identification of Continuous-time Errors-in-variables Models

Authors:	Mahata Kaushik, University of Newcastle, Australia Garnier Hugues, Universite Henri Poincare, Nancy, France
Topic:	1.1 Modelling, Identification & Signal Processing
Session:	Methods for Errors-in-Variables
Keywords:	Errors-in-variables, system identification, continuous-time models, linear estimation

Abstract

A novel direct approach for identifying continuous-time linear dynamicerrors-in-variables models is presented in this paper. The effects of the noise onthe state-variable filter outputs are analyzed. Subsequently, a search-free algorithmto obtain consistent continuous-time parameter estimates in the errors-in-variablesframework is derived. The performances of the proposed algorithm are illustratedwith some numerical simulation examples.