Estimating IDCurrent of 32nm FinFET by Artificial Neural Networks

Abstract

A large amount of MOSFET and CMOS transistors are produced day by day, and circuits are becoming very complex in this direction. As the circuit becomes more complex, it will be difficult to calculate the parameters of the transistor to be used in the circuit design, and this will make it difficult to determine the ID current in the circuits established with the transistor. To reduce the computation to a minimum, more transistors must be used, but this will result in larger chip usage. Fin field effect transistor (FinFET) technology is used to solve this problem. FinFET has a 3D transistor structure. Silicon fin on FinFET distinguishes it from other transistors. In this study, the closeness of the ID current obtained from the simulation program against the changing channel width and channel length parameters of FinFET technology by us was compared using Artificial Neural Networks. Error rates were obtained as a result of these comparison values. These error rates will give the circuit designer an idea of the physical properties of the element in the circuit with minimum time loss. © 2023 IEEE.