基于多层神经网络的HEMT大信号特性研究

A Large-Signal Characterization of an HEMT Using a Multilayered Neural Network

提出一种利用小信号等效电路模型,通过使用多层神经网络来描述高电子迁移率晶体管(HEMT)与偏置有关的大信号特性方法.通过实验发现,用一个2输入,5层28个神经元的神经网络,可以同时提取出7个本征参数.计算机仿真表明,该网络具有精度较高的特性以及很好的预测性.

We propose an approach to describe the large-signal behavior of a high electron-mobility transistor（HEMT） by using a multilayered neural network in terms of small-signal equivalent-circuit model.We successfully extract seven intrinsic elements with a two-input and five-layered neural network（composed of 28 neurons） with excellent accuracy and generates good extrapolations.