Premium
Spectral density based estimation of continuous‐time ARMAX process parameters
Author(s) -
Mossberg M.
Publication year - 2012
Publication title -
asian journal of control
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.769
H-Index - 53
eISSN - 1934-6093
pISSN - 1561-8625
DOI - 10.1002/asjc.338
Subject(s) - mathematics , discrete time and continuous time , polynomial , least squares function approximation , estimation theory , process (computing) , control theory (sociology) , mathematical optimization , computer science , algorithm , statistics , control (management) , mathematical analysis , estimator , artificial intelligence , operating system
The continuous‐time ARMAX model is a standard model that can be used for describing continuous‐time stochastic dynamic systems for control purposes. In this note, the problem of estimating the parameters in such a model from discrete‐time data is considered. In the proposed solution, the parameters in the denominator polynomial are estimated using a continuous‐time Yule‐Walker equation. Thereafter, the parameters in the numerator polynomials are estimated using an approach based on the spectral density of the output signal and regularized least squares. The method is sub‐optimal but easy to apply and the given estimates can be used directly or as initial values for the maximum likelihood method.Copyright © 2010 John Wiley and Sons Asia Pte Ltd and Chinese Automatic Control Society