Title :
Identification of nonlinear systems with evolving networks
Author :
Wasniowski, Richard A.
Author_Institution :
Sandia Res. Center, Albuquerque, NM, USA
Abstract :
This paper discusses and shows how computation intensive nonlinear identification problems can be computed efficiently using evolving networks and clusters of workstations. Simulations are conducted to study the performance of this approach with different nonlinear systems. Results of developing parallel algorithms for system identification are discussed
Keywords :
identification; neural nets; nonlinear systems; parallel algorithms; simulation; evolving network; identification; nonlinear system; parallel GMDH algorithm; simulation; workstation cluster; Computational modeling; Computer networks; Data handling; Nonlinear dynamical systems; Nonlinear systems; Parallel algorithms; Polynomials; Signal processing; System identification; Workstations;
Conference_Titel :
Circuits and Systems, 1996., IEEE 39th Midwest symposium on
Conference_Location :
Ames, IA
Print_ISBN :
0-7803-3636-4
DOI :
10.1109/MWSCAS.1996.592846
