Title :
A note on stochastic approximation algorithms in system identification
Author_Institution :
Drexel University, Philadelphia, PA, USA
fDate :
8/1/1972 12:00:00 AM
Abstract :
This correspondence considers the application of stochastic approximation algorithms to a broad class of system identification problems. Both asymptotic and initial convergence properties of the algorithms are discussed. A suboptimal procedure for parameter selection and a means of convergence acceleration are suggested.
Keywords :
Parameter identification; Stochastic approximation; Acceleration; Approximation algorithms; Convergence; Eigenvalues and eigenfunctions; Estimation error; Stochastic processes; Stochastic resonance; Stochastic systems; System identification; Transient response;
Journal_Title :
Automatic Control, IEEE Transactions on
DOI :
10.1109/TAC.1972.1100069
