Title :
Neural modelling of the large-signal drain current of the dual-gate MESFET with DC and pulsed I-V measurements
Author :
Abdeen, M. ; Yagoub, M.C.E.
Author_Institution :
Ottawa Univ., Ont., Canada
Abstract :
This work presents a neural network model of the drain current for dual-gate MESFET. The model is a combination of two sub-models; a static model represented by DC IV characteristics and dynamic model represented by pulsed IV characteristics. Pulsed measurements are performed at many bias points to capture the full dynamic behaviour of the device. The final model has a total of 25 neurons and is generated in a few minutes. The measurements and model data are in very good agreement with model error of less than 1%.
Keywords :
MESFET integrated circuits; neural chips; semiconductor device models; DC measurements; bias points; dual-gate MESFET; dynamic model; full dynamic behaviour; large-signal drain current; neural modelling; neural network model; pulsed I-V measurements; static model; Current measurement; Dispersion; Electrical resistance measurement; MESFETs; Microwave FETs; Microwave devices; Neural networks; Pulse measurements; Radio frequency; Transconductance;
Conference_Titel :
Circuits and Systems, 2004. ISCAS '04. Proceedings of the 2004 International Symposium on
Print_ISBN :
0-7803-8251-X
DOI :
10.1109/ISCAS.2004.1329928
