An ANN-Physical Hybrid GaN HEMT Model for 5G Power Amplifiers

Compared to other semiconductor technologies, GaN power amplifiers (PAs) demonstrate better efficiency, gain and thermal performance, which makes them very promising in the present 5G era. To design 5G GaN PAs, accurate and scalable GaN HEMT device modelling is very necessary. This paper proposes a hybrid physical GaN HEMTs model for 5G power amplifiers. Combine the concept of ASM-GaN-HEMT model and ANN technique, the simulation results of the proposed model in this paper are highly consistent with the test results in terms of the DC and RF characteristics. Multi-Objective Particle Swarm Optimization (MOPSO) is employed to derive the accurate surface potential values, which are the key parameters in ASM-GaN-HEMT. With the dimension information in the target equations, the derived GaN model will have certain scalability in device size. Also, artificial neural network (ANN) technique is utilized to build charge and nonlinear current models with the surface potential values. Effective self-heating and trapping effect models are introduced to demonstrate nonlinear effects. The proposed method significantly improves the accuracy of the S parameter compared to traditional models, reducing the large signal performance error to within 5%.