Title :
A new macromodeling approach for nonlinear microwave circuits based on recurrent neural networks
Author :
Fang, Y.H. ; Yagoub, M.C.E. ; Wang, F. ; Zhang, Q.J.
Author_Institution :
Dept. of Electron., Carleton Univ., Ottawa, Ont., Canada
Abstract :
For the first time, recurrent neural networks (RNN) are trained to learn the dynamic responses of nonlinear microwave circuits. Once trained, the RNN macromodel provides fast prediction of the full analog behavior of the original circuit and can be used for high level simulation and optimization.
Keywords :
analogue circuits; circuit optimisation; circuit simulation; microwave circuits; nonlinear network analysis; recurrent neural nets; RNN macromodel; dynamic responses; full analog behavior; high level optimization; high level simulation; macromodeling approach; nonlinear microwave circuits; recurrent neural networks; Circuit simulation; Circuit topology; Design optimization; Differential equations; Equivalent circuits; Microwave circuits; Microwave devices; Neural networks; Nonlinear circuits; Recurrent neural networks;
Conference_Titel :
Microwave Symposium Digest. 2000 IEEE MTT-S International
Conference_Location :
Boston, MA, USA
Print_ISBN :
0-7803-5687-X
DOI :
10.1109/MWSYM.2000.863321
