Title :
Nonparametric Identification of Weighting Function by Orthogonal Series Method
Author :
Rutkowski, Lossek
Author_Institution :
Technical University of Cz¿stochowa, Poland
Abstract :
A nonparametric algorithm for identifying the weighting function in linear dynamic continuous-time systems is proposed. The algorithm is derived from orthogonal series estimates of derivatives of regression functions. Asymptotic properties are investigated for a general orthogonal system and sufficient conditions for convergence in the sense of the mean square error are given.
Keywords :
Automatic control; Convergence; Diffusion processes; Instruction sets; Mean square error methods; Noise measurement; Process control; Sequences; Sufficient conditions; System identification;
Conference_Titel :
American Control Conference, 1982
Conference_Location :
Arlington, VA, USA
