Title :
Optimal estimation of operator-valued stochastic processes and applications to distributed parameter systems
Author_Institution :
Michigan State University, East Lansing, Michigan
Abstract :
The paper develops optimal estimation equations for operator-valued discrete-time wide-sense Markov processes. The signals are viewed as linear transformations of wide-sense martingale processes, a general representation which yields relatively simple estimates and error covariances. The infinite-dimensional results are applied to prediction, filtering and smoothing in distributed parameter systems.
Keywords :
Distributed parameter systems; Equations; Filtering; Hilbert space; Markov processes; Power engineering and energy; Signal processing; Smoothing methods; State estimation; Stochastic processes;
Conference_Titel :
Decision and Control, 1972 and 11th Symposium on Adaptive Processes. Proceedings of the 1972 IEEE Conference on
Conference_Location :
New Orleans, Louisiana, USA
DOI :
10.1109/CDC.1972.268950