Non-iterative optimal min-max instrumental variable method for system identification

Petre Stoica, Arye Nehorai

Research output: Contribution to journalArticlepeer-review

2 Scopus citations

Abstract

An optimal min-max instrumental variable (IV) method for estimating the parameters in difference equation models is presented. The method gives the smallest estimation error variance in the ‘worst noise’ case from a prespecified class of noises. Implementation of this new method does not require knowledge of system or noise parameters and therefore can be done without iteration. This is an attractive feature since other min-max or generally optimal IV methods previously introduced require at least a knowledge of system parameters, and cannot be implemented directly. Results of simulations demonstrate the applicability of the method to both artificial and real data.

Original languageEnglish
Pages (from-to)1759-1769
Number of pages11
JournalInternational Journal of Control
Volume47
Issue number6
DOIs
StatePublished - Jun 1988

Fingerprint

Dive into the research topics of 'Non-iterative optimal min-max instrumental variable method for system identification'. Together they form a unique fingerprint.

Cite this