Waveforms-based large-signal identification of transistor models

Measurements of low- and high-frequency vector-calibrated large-signal waveforms are exploited in this work to identify the parameters of a FET nonlinear model. The IDS nonlinear current source and the nonlinear charge sources' parameters are respectively determined from a small set of low- (2 MHz) and high-frequency (8 GHz) load-pull measurements by using a least square numerical optimization. Under low-frequency operation the contribution of the charge sources and any other reactive element can be neglected. In this way the identification of the IDS parameters is more accurate while remarkably speeding up the optimization routine as well. The proposed procedure is quite general and can be applied to different types of active devices. As case study, a 0.25-µm GaAs pHEMT is considered and the extracted model is validated under conditions different than the ones exploited within the identification step. A very good agreement between model predictions and experimental data is achieved.