Title :
A large-signal neural network model for the dual gate MESFET
Author :
Abdeen, M. ; Bennett, J. ; Yagoub, M.C.E.
Author_Institution :
SITE, Ottawa Univ., Ont., Canada
Abstract :
The paper presents for the first time a large-signal neural model of the dual-gate MESFET. The neural model is a five-layer perceptrons (MLP5) with one input, one output, and three hidden layers. The model is based on pulsed I-V measurements to better represent the RF device behavior and to neutralize the effect of channel heating on the model accuracy. The neural model is shown to be in an excellent agreement with measurements (with an error less than 1%).
Keywords :
Schottky gate field effect transistors; circuit CAD; perceptrons; RF device; channel heating; dual gate MESFET; five-layer perceptrons; neural network model; pulsed I-V measurements; Current measurement; Dispersion; Electrical resistance measurement; FETs; MESFETs; Microwave devices; Neural networks; Pulse measurements; Radio frequency; Transconductance;
Conference_Titel :
Circuits and Systems, 2003 IEEE 46th Midwest Symposium on
Print_ISBN :
0-7803-8294-3
DOI :
10.1109/MWSCAS.2003.1562539
