Modeling of subthreshold characteristics for double gate MOSFET using neural networks and genetic algorithm

As the metal-oxide-semiconductor field-effect transistor (MOSFET) technology has been developed, the short-channel effects become significant. To overcome these limitations, double gate (DG) MOSFET has been considered and predicting the device characteristics according to device parameters has been important. In this paper, we present the neural networks (NNET) modeling methodology to predict subthreshold characteristics such as threshold voltage (V_TH) and subthreshold swing (S_SUB) for DG MOSFET. After the NNET model is established, the genetic algorithm (GA) is used to find the device parameters' design space.