Title :
A robust algorithm for random parameter tracking
Author :
Juditsky, Anatoli ; Priouret, Pierre
Author_Institution :
IRISA, Rennes, France
fDate :
6/1/1994 12:00:00 AM
Abstract :
An algorithm for tracking time-varying parameters is studied. It is shown to be exponentially stable independently of the choice of algorithm parameters. We derive some useful nonasymptotic estimates for the tracking error and develop the asymptotically optimal version of the method
Keywords :
error statistics; parameter estimation; probability; stochastic processes; time-varying systems; error equations; nonasymptotic approximations; nonasymptotic estimates; probability space; random parameter tracking; time varying parameters; tracking error; Adaptive equalizers; Convergence; Covariance matrix; Phase locked loops; Random variables; Resonance light scattering; Robustness; Signal processing; Signal processing algorithms; Stochastic resonance;
Journal_Title :
Automatic Control, IEEE Transactions on
