Abstract
This work deals with the detection of minimal order for linear systems with piecewise constant coefficients. In this case, it is impossible to use off-line algorithms. We propose, here, a real-time method able to detect the minimal order of a linear system. The basic mathematic tool is given by a simple divisibility condition of the charateristic polynomial of the differential equation for stationary systems. This necessary and sufficient condition is used to define an algorithm able to give the reduced equation. Then, we show that this result can be directly extended to linear piecewise stationary systems for which we search a change if the order. The real-time method that we obtain, is checked in simulation on deterministic systems and on systems with measured noisy informations.
Membre du GRECO Traitement du Signal et des Images
Preview
Unable to display preview. Download preview PDF.
VII; Bibliography
— R.E. KALMAN—P.L. FALB—H.A. ARBIB: Topics in mathematical systems theory Mac Graw 1970
— G.ALENGRIN—G. FAVIER: New stochastic realisation algorithms for identification of ARMA models IEEE ASSP conf. TUSLA 1978
— L. MARPLE: Digital spectral analysis with applications Prentice Hall 1986
— J.M. HILL: Solution of differential equations by means of one parameter group Pitman Research notes in Math. 1982.
— E. KAMEN: The poles and zeros of a linear time-invariant system Linear algebra and its application 1988–98 PP 263–289
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1990 Springer-Verlag
About this paper
Cite this paper
Guglielmi, M. (1990). On-line detection of minimal order for linear piecewise stationary systems. In: Bensoussan, A., Lions, J.L. (eds) Analysis and Optimization of Systes. Lecture Notes in Control and Information Sciences, vol 144. Springer, Berlin, Heidelberg. https://doi.org/10.1007/BFb0120072
Download citation
DOI: https://doi.org/10.1007/BFb0120072
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-52630-8
Online ISBN: 978-3-540-47085-4
eBook Packages: Springer Book Archive