Multiobjective FET modeling using particle swarm optimization based on scattering parameters with Pareto optimal analysis

In this paper, design-oriented field effect transistor (FET) models are produced. For this purpose, FET modeling is put forward as a constrained, multiobjective optimization problem. Two novel methods for multiobjective optimization are employed: particle swarm optimization (PSO) uses the single-objective function, which gathers all of the objectives as aggregating functions; and the nondominated sorting genetic algorithm-II (NSGA-II) sorts all of the trade-off solutions on the Pareto frontiers. The PSO solution is compared with the Pareto optimum solutions in the biobjective plane and the success of the first method is verified. Furthermore, the resulting FET models are compared with similar FET models from the literature, and thus a comparative study is put forward with respect to the success of the optimization algorithms and the performances and utilizations of the models in the amplification circuits