Title :
Persistency of excitation, identification, and radial basis functions
Author :
Kurdila, Andrew J. ; Narcowich, Francis J. ; Ward, Joseph D.
Author_Institution :
Dept. of Aerosp. Eng., Texas A&M Univ., College Station, TX, USA
Abstract :
Discusses identification algorithms whose convergence and rate of convergence hinge on the regressor vector being persistently exciting. The authors then show that if the regressor vector is constructed out of radial-basis-function approximants, it will be persistently exciting, provided a kind of “ergodic” condition is satisfied. In addition, the authors provide bounds on parameters associated with the persistently exciting regressor vector; these parameters are connected both with the convergence and rates of convergence of the algorithms involved
Keywords :
convergence; feedforward neural nets; identification; ergodic condition; identification; persistency of excitation; radial basis functions; rate of convergence; regressor vector; Adaptive systems; Aerospace engineering; Approximation methods; Convergence; Fasteners; Least squares approximation; Mathematics; Neural networks; Scattering; Sufficient conditions;
Conference_Titel :
Decision and Control, 1994., Proceedings of the 33rd IEEE Conference on
Conference_Location :
Lake Buena Vista, FL
Print_ISBN :
0-7803-1968-0
DOI :
10.1109/CDC.1994.411479