Title :
Parameter and state estimation of a nonstationary process
Author :
Furuta, K. ; Paquet, J.-G.
Author_Institution :
Tokyo Institute of Technology, Tokyo, Japan
fDate :
12/1/1969 12:00:00 AM
Abstract :
A class of nonstationary processes which changes dynamic characteristics in a discrete Markovian way and a method to identify a parameter and to estimate a state variable is proposed. This method is based on the mean-square error criterion using the conditional probability approach under the assumption that the conditional probability distributions of both the parameter and the state variable are Gaussian.
Keywords :
Markov processes; Nonstationary stochastic processes; Parameter estimation; State estimation; Electronic switching systems; Equations; Laboratories; Linear systems; Parameter estimation; Random variables; State estimation; Statistics; Stochastic processes; Telephony;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.1969.1099327
