A novel piecewise model for pHEMTs with accurate Ids and its first three derivatives (gm, gm2 and gm3) is presented. The entire operating region is divided into several subregions. Aiming at improving the model accuracy in each subregion, the conventional Angelov models are developed with different optimized parameters. To solve the problem of the discontinuity between adjacent subregions, the artificial neural network (ANN) technique is utilized. The validity of the proposed model is demonstrated by the comparison of measured and modeled results for a pHEMT.

Additional Metadata
Keywords artificial neural network, high-order derivative, Nonlinear modeling, pHEMT modeling, piecewise method
Persistent URL dx.doi.org/10.1109/NEMO.2016.7561614
Conference 2016 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2016
Citation
Zhu, Y.-Y. (Yuan-Yuan), Fu, H.-P. (Hai-Peng), Zhang, Q.J, & Ma, J.-G. (Jian-Guo). (2016). A novel model for pHEMTs with accurate high-order derivatives. In 2016 IEEE MTT-S International Conference on Numerical Electromagnetic and Multiphysics Modeling and Optimization, NEMO 2016. doi:10.1109/NEMO.2016.7561614