A brief review over neural network modeling techniques
Mohammad Reza Mohammadi, Sayed Alireza Sadrossadat, Mir Gholamreza Mortazavi, Behzad Nouri · 2017 IEEE International Conference on Power, Control, Signals and Instrumentation Engineering (ICPCSI) · 2017
Artificial neural networks (ANN) have been recently emerged as a powerful computer-aided design (CAD) tool for modeling devices and circuits. The overall objective of this paper is to do a survey over different neural network techniques for both frequency domain and transient modeling of circuits and components. The static models discussed in this paper are multilayer perceptron (MLP) and Radial basis function (RBF) neural networks which are mostly used to model and analyze the frequency domain behavior of the circuits. On the other hand, recurrent neural networks (RNN) and dynamic neural network (DNN) that are considered to be time-domain ANNs are discussed. These neural networks permit modeling and analyzing the transient behavior of the nonlinear circuits/components.