This article reports a comparative study of two artificial neural network structures and associated variants used to describe and predict the behavior of 2 × 200 μm2 GaN high electron mobility transistors (HEMTs), utilizing radiofrequency characterization. Two architectures namely multilayer perceptron and cascade feedforward, have been investigated in this work to develop the behavioral model. A study is conducted utilizing the two architectures, all trained using Levenberg‐Marquardt, in terms of accuracy, convergence rate, and generalization capability to develop the behavioral model of GaN HEMT. However, to ensure the robustness of the model, accuracy, convergence rate, time elapsed, and generalization capability of the proposed model is also tested under couple of training algorithms, activation functions, number of hidden layers and neuron embedded inside it, methods for initialization of weights and bias and certain other vital parameters playing vital role in influencing the model accuracy and effectiveness. An excellent agreement found between measured S‐parameters and the proposed model proves the effectiveness of the proposed approach and excellent prediction ability for a sweeping multibias set and broad frequency range of 1 to 18 GHz. Moreover, a very good generalization capability is also recorded under variation of crucial parameters of GaN HEMT‐based neural model.
Summary
In this paper, development of a small signal model for 2 × 200 μm GaN HEMT based on the conventional 20‐element model is presented. The proposed model presents a direct parameter extraction algorithm, instead of the hybrid optimization approach, that provides simplification, accuracy, and less computational complexity. The extrinsic elements are extracted using a modified cold pinch‐off condition while discarding the unwanted forward biasing of the gate. The negative drain to source capacitance Cds is also observed in the ohmic region (for smaller VDS). An excellent agreement found between the measured and modeled data for a wide range of frequencies and bias values shows the effectiveness of the proposed approach. The proposed modeling technique is validated with a good agreement between the achieved bias dependency of intrinsic parameter values and the respective theoretical parameter values.
The work reported in this paper explores a novel Particle Swarm Optimization (PSO) tuned Support Vector Regression (SVR) based technique to develop the small-signal behavioral model for GaN High Electron Mobility Transistor (HEMT). The proposed technique investigates issues such as kernel selection and model optimization usually encountered in the application of SVR to model the GaN based HEMT devices. Here, the PSO algorithm is utilized to find the optimal hyperparameters to minimize the fitness function. To enumerate the efficiency and the generalization capability of the predictors, the performance of the model is investigated in terms of mean square error (MSE) and mean relative error (MRE). A very good agreement is found between the measured S-parameters and the proposed model for multi-biasing sets over the complete frequency range of 1 GHz-18 GHz. The proposed technique is even used to test the frequency extrapolation capability of the model. A comparative analysis indicates that the proposed PSO-SVR predictor achieves significantly improved computational efficiency and the overall prediction accuracy. To demonstrate the ready usefulness of the modeling approach, the developed model has been incorporated in CAD environment using MATLAB Cosimulation in ADS Ptolemy. Subsequently, the small-signal stability analysis is performed and gain of a power amplifier configuration designed using the proposed GaN HEMT model is determined.
This study employs support vector regression (SVR) to develop an accurate and reliable intrinsic parameter extraction model for gallium nitride (GaN) high electron mobility transistors (HEMT) using two different geometries of 2 × 200 μm and 4 × 100 µm. The key aspect of the proposed approach is the use of nonlinear Gaussian kernel to transform the input space into a high‐dimensional feature space. It then allows the application of learning technique to develop a reliable procedure for parameter extraction. The proposed extraction model of GaN HEMT has been developed for a broad range of frequency, from 1 to 18 GHz, with multi‐biasing sets for HEMTs of two different geometries. Moreover, the proposed model is made scalable in terms of geometry parameters and therefore can be used to predict the intrinsic parameters and enumerate scaling efficiency of GaN HEMTs by investigating the geometry parameters. A good agreement is observed between the measured S‐parameters and the proposed model for the complete frequency range. It is shown that the proposed approach is simple, novel and can be readily incorporated into computer‐aided design tool for an accurate and expedited design process of RF and microwave circuits.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.