The paper proposes a new frequency domain method for identifying linear dynamic errors-in-variables (EIV) models. The noise-free input is an arbitrary signal, not necessarily periodic and the input and output noises are additive and uncorrelated white processes. The method combines, in a frequency domain context, the characteristics of the Frisch scheme and the properties of the Yule-Walker equations. The features of the method are illustrated by means of numerical examples.
Frequency domain EIV identification combining the Frisch scheme and Yule-Walker equations / Soverini, Umberto; Soderstrom, Torsten. - ELETTRONICO. - (2015), pp. 7330839.2038-7330839.2043. (Intervento presentato al convegno European Control Conference, ECC 2015 tenutosi a Linz, Austria nel 2015) [10.1109/ECC.2015.7330839].
Frequency domain EIV identification combining the Frisch scheme and Yule-Walker equations
SOVERINI, UMBERTO;
2015
Abstract
The paper proposes a new frequency domain method for identifying linear dynamic errors-in-variables (EIV) models. The noise-free input is an arbitrary signal, not necessarily periodic and the input and output noises are additive and uncorrelated white processes. The method combines, in a frequency domain context, the characteristics of the Frisch scheme and the properties of the Yule-Walker equations. The features of the method are illustrated by means of numerical examples.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.