Large-signal modeling of on-wafer microwave transistors based on response surface methodology

In this paper we extract the large-signal model of a microwave transistor by using the response surface methodology. This behavioral modeling approach combines design of experiments and automated model extraction. As case study we considered an on-wafer 0.15 μm GaAs pHEMT. The proposed model is compared with well-established Chalmers model. We show that a high level of accuracy can be achieved with response surface methodology. However, the methodology is computationally complex.