A Multi-Bias Correlated Parameter Extraction Method for Microwave Transistor Modeling

This paper proposes a new multi-bias correlated parameter extraction method for microwave transistor modeling. The initial small-signal equivalent circuit element values are extracted using an analytical method. Optimum element values are obtained using a gradient optimizer by including multiple sets of different DC bias models in one fitting process. The validity of the developed modeling approach has been verified by the good agreement and less than 2% deviation between the simulated and measured results for different sizes of gallium nitride transistors over a wide range of DC bias conditions and broad frequency range.