Title :
Application of neural networks in microwave circuit modelling
Author :
Peik, S.F. ; Coutts, G. ; Mansour, R.R.
Author_Institution :
COM DEV, Cambridge, Ont., Canada
Abstract :
Multi-layer perceptron neural networks have been developed for resonant microwave structures with highly non-linear responses. The network training data is compiled by an accurate electromagnetic simulation of the circuits. Two numerical examples are given: a microstrip patch antenna and a planar 3-pole filter. Utilizing the trained networks the circuits have been also optimized to certain specifications
Keywords :
circuit analysis computing; circuit optimisation; learning (artificial intelligence); microstrip antennas; microwave antennas; microwave filters; multilayer perceptrons; nonlinear network analysis; poles and zeros; resonance; circuit optimization; electromagnetic simulation; microstrip patch antenna; microwave circuit modelling; multi-layer perceptron; network training data; neural networks; nonlinear responses; planar 3-pole filter; resonant microwave structures; trained networks; Circuit simulation; Microstrip antennas; Microstrip filters; Microwave circuits; Multi-layer neural network; Multilayer perceptrons; Neural networks; Patch antennas; Resonance; Training data;
Conference_Titel :
Electrical and Computer Engineering, 1998. IEEE Canadian Conference on
Conference_Location :
Waterloo, Ont.
Print_ISBN :
0-7803-4314-X
DOI :
10.1109/CCECE.1998.685651
