distinguishability conditions for stationary linear systems: Distinguishability conditions for stationary linear systems

Uloženo v:
Podrobná bibliografie
Název: distinguishability conditions for stationary linear systems: Distinguishability conditions for stationary linear systems
Autoři: A. A. Lomov
Zdroj: Differential Equations. 39(2):283-288
Informace o vydavateli: Springer US, New York, NY; Pleiades Publishing, New York, NY; MAIK ``Nauka/Interperiodica'', Moscow, 2003.
Rok vydání: 2003
Témata: Realizations from input-output data, equivalence of parameters, System identification, identifiability, distinguishability, realizations from input-output data, ARMA models, system identification
Popis: Vector ARMA models are considered where the coefficient matrices are functions of the same parameter vector within a given subset of the \(n\)-dimensional real vector space. Distinguishability is introduced as a necessary condition for the identifiability of parameters and is rigorously defined; its interpretation is related to the equivalence of parameters giving rise to the same input/output pairs in time. Then, necessary and sufficient conditions are given for local and global distinguishability.
Druh dokumentu: Article
Popis souboru: application/xml
ISSN: 0012-2661
DOI: 10.1023/a:1025117418970
Přístupová URL adresa: https://zbmath.org/2146341
https://doi.org/10.1023/a:1025117418970
https://link.springer.com/article/10.1023/A:1025117418970
https://link.springer.com/content/pdf/10.1023/A:1025117418970.pdf
Přístupové číslo: edsair.dedup.wf.002..2e1c86d9fb9ed2f64ffe42369894e5ad
Databáze: OpenAIRE
Popis
Abstrakt:Vector ARMA models are considered where the coefficient matrices are functions of the same parameter vector within a given subset of the \(n\)-dimensional real vector space. Distinguishability is introduced as a necessary condition for the identifiability of parameters and is rigorously defined; its interpretation is related to the equivalence of parameters giving rise to the same input/output pairs in time. Then, necessary and sufficient conditions are given for local and global distinguishability.
ISSN:00122661
DOI:10.1023/a:1025117418970