A cumulant statistics-based method for continuous-time errors-in-variables model identification - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2006

A cumulant statistics-based method for continuous-time errors-in-variables model identification

Hugues Garnier
  • Fonction : Auteur
  • PersonId : 836179
  • IdRef : 120561565
Marion Gilson

Résumé

In this paper, the problem of identifying stochastic linear continuous-time systems from noisy input/output data is addressed. The input of the system is assumed to have a skewed probability density function, whereas the noises contaminating the data are assumed to be symmetrically distributed. The third-order (cross-) cumulants of the input/output data are then (asymptotically) insensitive to the noises, that can be colored and/or mutually correlated. Using this noise-cancellation property two computationally simple estimators are proposed. The usefulness of the proposed algorithms is assessed through a numerical simulation.
Fichier principal
Vignette du fichier
WorkshopTLSEIV2006HAL_Thil.pdf (52.59 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-00121389 , version 1 (20-12-2006)

Identifiants

  • HAL Id : hal-00121389 , version 1

Citer

Stéphane Thil, Hugues Garnier, Marion Gilson. A cumulant statistics-based method for continuous-time errors-in-variables model identification. 4th International Workshop on Total Least Squares and Errors-in-variables Modeling, Aug 2006, Leuven, Belgium. ⟨hal-00121389⟩
58 Consultations
46 Téléchargements

Partager

Gmail Facebook X LinkedIn More