Title: Adaptive modelling of non-linear errors-in-variables systems

Authors: Yuan Li; Zheng Qin; Laixiang Shan

Addresses: School of Software, Tsinghua University, Beijing 100084, China ' School of Software, Tsinghua University, Beijing 100084, China ' School of Software, Tsinghua University, Beijing 100084, China

Abstract: This paper discusses the adaptive modelling algorithm of the non-linear Errors-in-Variables (EIV) system. The non-linear EIV model is designed as a Wiener model with noisy input and noisy output measurements. The model consists of a moving average model linear subsystem and a piece-wise linear function as the non-linear subsystem. In order to ensure the online real-time updating of system parameters, the recursive form is designed for the algorithm to keep its adaptive properties and improve the parameters' identification efficiency. The derivation process of the algorithm ensures the recursive calculation results converge to their corresponding true values. The validity and convergence of the algorithm is confirmed by a simulation experiment with the help of Matlab.

Keywords: errors-in-variables system; adaptive modelling; nonlinear EIV modelling; Wiener model; recursive algorithm; simulation.

DOI: 10.1504/IJCAT.2016.073604

International Journal of Computer Applications in Technology, 2016 Vol.53 No.1, pp.71 - 81

Received: 15 Jan 2014
Accepted: 15 May 2014

Published online: 13 Dec 2015 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article