Large-Signal Modeling of GaN HEMTs Using Hybrid GA-ANN, PSO-SVR, and GPR-Based Approaches

Anwar Jarndal,Fadhel M Ghannouchi,Mohammad Hashmi,Saddam Husain

doi:10.1109/jeds.2020.3035628

Abstract

This article presents an extensive study and demonstration of efficient electrothermal large-signal GaN HEMT modeling approaches based on combined techniques of Genetic Algorithm (GA) with Artificial Neural Networks (ANN), and Particle Swarm optimization (PSO) with Support Vector Regression (SVR). Another promising Gaussian Process Regression (GPR) based large-signal modeling approach is also explored and presented. The GA-ANN addresses the typical problem of local minima associated with the backpropagation (BP) based ANN. The GA successfully aids in the determination of optimal initial values for BP-ANN and enables it to find a unique optimal solution after subsequent of iterations with higher rate of convergence. This is also achieved using PSO-SVR with lower optimization variables. The developed modeling techniques are demonstrated and used to simulate the gate and drain currents of a 2-mm GaN device. All the models are relatively simple, practical, and easy to implement. The gate and drain currents models are embedded in an equivalent large-signal circuit's model and built in Advanced Design System (ADS) software. The implemented model is validated by large-signal measurements and very good fitting results have been obtained. The model also showed an accurate simulation for a nonlinear power amplifier with very good computational speed and convergence.

Highlights

The power amplifiers employed in broadcasting and communication transmitter applications are high power and are intrinsically non-linear [1]–[7]
The reliability of power amplifiers in wireless and broadcasting transmitters depends on the accuracy of the employed GaN High Electron Mobility Transistor (HEMT) device largesignal models [14]
Increasing the order of the Artificial Neural Networks (ANN) to improve the model fitting complicates the model implementation in CAD software and affect the convergence of simulation. This aspect can be addressed by utilizing global optimization techniques such as genetic algorithm (GA) and particle swarm optimization (PSO) as they have been found to be very good alternatives to train neural networks [34], [35]

Summary

INTRODUCTION

The power amplifiers employed in broadcasting and communication transmitter applications are high power and are intrinsically non-linear [1]–[7]. The black box nature of this technique reduces the cost of searching for proper formula and computation of the model’s parameters It exhibits higher rate of convergence when compared to the analytical modeling and its prediction capability can be improved by choosing suitable model topology and activation function. Increasing the order of the ANN to improve the model fitting complicates the model implementation in CAD software and affect the convergence of simulation This aspect can be addressed by utilizing global optimization techniques such as genetic algorithm (GA) and particle swarm optimization (PSO) as they have been found to be very good alternatives to train neural networks [34], [35].

EQUIVALENT CIRCUIT MODEL

PSO AUGMENTED SVR BASED MODEL

GPR BASED MODEL

THE DEVELOPED MODEL VALIDATION

Findings

VIII. CONCLUSION