In this paper we study the positivity of an estimated time series (or impulse response) described by a sum of real exponentials. Such a problem, i.e. fitting a time series as a sum of exponential functions, is a longstanding one and has been studied by a very large number of authors. Positivity of the time series is often required in many diverse fields of application where it is a direct consequence of the physics underlying the process under study. In this paper we suggest a possible way to incorporate the positivity constraints in the parameters estimation process. Copyright © 2002 IFAC
Positivity of multi-exponential models / Farina, Lorenzo; P., Ruisi. - 15:(2002), pp. 53-57. (Intervento presentato al convegno 15th World Congress of the International Federation of Automatic Control tenutosi a Barcelona; Spain nel 21-26 July 2002).
Positivity of multi-exponential models
FARINA, Lorenzo;
2002
Abstract
In this paper we study the positivity of an estimated time series (or impulse response) described by a sum of real exponentials. Such a problem, i.e. fitting a time series as a sum of exponential functions, is a longstanding one and has been studied by a very large number of authors. Positivity of the time series is often required in many diverse fields of application where it is a direct consequence of the physics underlying the process under study. In this paper we suggest a possible way to incorporate the positivity constraints in the parameters estimation process. Copyright © 2002 IFACI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
