| Title |
Artificial Neural Network Model for Predicting Work Function Variation (WFV) and Random Dopant Fluctuation (RDF)-induced Variation of Electrical Characteristics of 5 nm Node FinFET |
| Authors |
김지환(Jihwan Kim) ; 이재준(Jaejoon Lee) ; 이진웅(Jinwoong Lee) ; 임재혁(Jaehyuk Lim) ; 신창환(Changhwan Shin) |
| DOI |
https://doi.org/10.5573/ieie.2022.59.8.115 |
| Keywords |
Random dopant fluctuation(RDF); Work function variation(WFV); Artificial Neural Network(ANN); FinFET; Semiconductor Devices |
| Abstract |
Work function variation (WFV) and Random dopant fluctuation (RDF) are the main sources of Process-induced variation in Fin-shaped field effect transistor (FinFET). In this work, we proposed and developed the Artificial neural network (ANN) model for predicting the WFV/RDF-induced variation of electrical characteristics of 5nm node FinFET. This ANN model uses four input features [i.e., Average grain size(AGS), Source/Drain doping density(S/D doping), Retrograde channel doping concentration(RCD doping), Retrograde channel doping peak point(RCD Peak)] to predict 7 output features [i.e., off-state leakage current(), saturation drain current(), linear drain current(), low drain current(), high drain current(), saturation threshold voltage() and linear threshold voltage()] representing device performance. |