Title :
System identification using neural networks
Author :
Sugimoto, S. ; Matsumoto, M. ; Kabe, M.
Author_Institution :
Dept. of Electr. Eng., Ritsumeikan Univ., Kyoto, Japan
Abstract :
An identification problem for multivariable discrete-time linear systems is considered in the framework of neural networks and computing. A parallel computing algorithm for identification for discrete-time linear systems based on the modified Hopfield model with continuous states is proposed. The proposed parallel algorithm inherently can solve a wide variety of least-squares problems. Simulation results are presented
Keywords :
discrete time systems; identification; least squares approximations; linear systems; multivariable systems; parallel algorithms; identification; least-squares problems; modified Hopfield model; multivariable discrete-time linear systems; neural networks; parallel computing algorithm; Neural networks; Neurons; System identification;
Conference_Titel :
Systems Engineering, 1992., IEEE International Conference on
Conference_Location :
Kobe
Print_ISBN :
0-7803-0734-8
DOI :
10.1109/ICSYSE.1992.236938
