基于神经网络的GaAs微波及高速集成电路CAD技术

Feedforward Neural Network Based GaAs Microwave and VHSIC CAD

介绍了基于神经网络的 Ga As微波与高速电路 CAD方面的研究与开发工作 ,其中包括用神经网络进行器件与电路建模的原理与方法及这一模型在优化与统计分析中的应用。

This paper presents the researches and developments on the GaAs microwave and VHSIC CAD based upon the feedforward neural network modelling techniques in the GaAs IC CAD project.The general device and circuit performance modelling techniques using feedforward neural networks are described.Sampling scheme to generate the modelling data and training algorithm for the neural networks are discussed.Some examples are given to demonstrate the effectiveness of this technique.