A novel extraction approach of extrinsic and intrinsic parameters of InGaAs/GaN pHEMTs

A novel extraction approach of extrinsic and intrinsic parameters of InGaAs/GaN pHEMTs is presented, for the first time, artificial bee colony algorithm is applied to the global-optimization based parameter extraction and a novel intrinsic Y-parameter error function is proposed, which is not sensitive to noise and measurement uncertainty, and also highly consistent to the overall S-parameter error function. The extrinsic elements are optimized at multi-bias points and the intrinsic ones at specific bias points. Only broad ranges of extrinsic parameters are required, and the extraction is unique and accurate. This method has been verified by 2 × 75 and 4 ×100 um gate width InGaAs pHEMTs from 1 GHz to 40 GHz. Excellent agreement is achieved between the measured and modeled S-parameters at all the biases and the conservation of the gate charge is well satisfied which further validates this novel extraction method.