Artificial neural networks in extraction of equivalent circuit elements of microwave FETs

A procedure for extraction of equivalent circuit elements of microwave FETs using artificial neural networks is presented in this paper. A neural network is trained to predict equivalent circuit elements for given frequency and scattering parameters, thus avoiding optimization procedures in microwave circuit simulator. The neural model is developed for a wide device temperature range. An investigation concerning number of measured data necessary for development of reliable neural models as well as frequency influence on obtained values of equivalent circuit elements is done.