A Neural Network based Fast Parameter Extraction of Compact Hot Carrier Degradation Model in FinFETs

This study comprehensively investigates the parameter extraction flows for Hot Carrier Degradation (HCD) in advanced technology based on Neural Network (NN). The emerging NN-based approach and the conventional particle swarm algorithm are applied for the extraction of HCD-related parameters in the BSIM-CMG model, respectively. As verified by 16/14nm FinFET data, HCD-induced degraded characteristics can be well extracted. Taking the PSO algorithm as the baseline, the accuracy and efficiency of the NN-based approach to parameter extraction are comprehensively studied and compared. Based on the parameter extraction results and the computational costs, the emerging NN-based methods are considered to be more effective for applications in test data-intensive scenarios.