Novel ESD Compact Modeling Methodology Using Machine Learning Techniques for Snapback and Non-Snapback ESD Devices

Abstract

Novel ESD compact modeling methodology using machine learning techniques is proposed for the first time in this paper. ESD compact modeling for snapback and non-snapback ESD protection devices are discussed. Comparisons between conventional, novel fitting methodology and different machine learning models are discussed. The advantages of this novel methodology are introduced and methods to further improve the model accuracy are also discussed.