scientific article; zbMATH DE number 3922504

From MaRDI portal
Publication:3697004





System identification is a process of estimating the best values of unknown parameters from experimental data for constructing a model of a dynamic system. System identification is important in control design of a dynamic system, particularly in the design of an adaptive control system. The most widely used criteria for system identification are the least- squares estimate, best linear unbiased estimate and the maximum likelihood estimate. Numerous literatures on system identification are available for continuous as well as discrete-time systems. The paper deals specifically with the least-squares identification method. It presents a theorem on convergence and on the strong consistency of the least-squares method for identifying the parameters of a discrete-time system governed by a difference equation. Comparing to previous results it requires weaker conditions on noise and data needed for identification. It is pointed out that the result in \textit{L. Ljung}'s paper [see IEEE Trans. Autom. Control AC-21, 779-781 (1976; Zbl 0334.93048)] can be considered as a special case of the results in this paper. The paper is an interesting brief technical note.




Cited in
(19)








This page was built for publication:

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3697004)