Title :
Dynamic representation and recursive estimation of cyclic and two-dimensional processes
Author :
Strintzis, Michael G.
Author_Institution :
University of Pittsburgh, Pittsburgh, PA, USA
fDate :
10/1/1978 12:00:00 AM
Abstract :
Autoregressive-moving average dynamic models are construtted for nonstationary processes which exhibit cyclic behavior. Recursive linear mean-square prediction and filtering policies are then formulated which utilize these models. Application to a problem in two-dimensional image restoration is considered in detail.
Keywords :
Autoregressive moving-average processes; Nonstationary stochastic processes; Parameter estimation; Recursive estimation; Covariance matrix; Image restoration; Multiplexing; Power system modeling; Power system restoration; Predictive models; Recursive estimation; Signal processing; Tin; Vectors;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.1978.1101859
