Title :
The relationship between state space subspace identification methods and the EM method
Author :
Gibson, Stuart ; Ninness, Brett
Author_Institution :
Dept. of Electr. & Comput. Eng., Newcastle Univ., NSW, Australia
Abstract :
This paper exposes a close connection between subspace-type estimates and single iterates of the expectation maximisation method for maximum likelihood (ML) estimation. A key implication of this is that it suggests a means by which ML estimates for MIMO systems may be computed in a numerically robust and straightforward manner
Keywords :
MIMO systems; least squares approximations; maximum likelihood estimation; state-space methods; EM algorithm; MIMO systems; expectation maximisation method; maximum likelihood estimation; parameter estimation; state space; subspace identification; Kalman filters; MIMO; Maximum likelihood estimation; Parameter estimation; State estimation; State-space methods; Stochastic processes; Technological innovation; White noise; Zinc;
Conference_Titel :
Decision and Control, 2000. Proceedings of the 39th IEEE Conference on
Conference_Location :
Sydney, NSW
Print_ISBN :
0-7803-6638-7
DOI :
10.1109/CDC.2000.914161
