Machine Learning-Based Device Modeling and Performance Optimization for FinFETs

Article Engineering, Electrical & Electronic

Rapid MOSFET Contact Resistance Extraction From Circuit Using SPICE-Augmented Machine Learning Without Feature Extraction

Thomas Lu et al.

Summary: The article discusses a method for extracting defect features from IC electrical characteristics through dimensionality reduction, training a machine using simulated data, and validating the accuracy through experimental data testing.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

TCAD-Augmented Machine Learning With and Without Domain Expertise

Harsaroop Dhillon et al.

Summary: This article demonstrates the cost-effectiveness of using TCAD to generate data for building machine learning models for semiconductor devices. By training machines with TCAD I-V data, effective Schottky metal work functions and ambient temperatures of experimental devices can be deduced. Both machine learning models with and without domain expertise showed reasonable prediction accuracy despite noise and variations in the experimental device.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Machine Learning Aided Device Simulation of Work Function Fluctuation for Multichannel Gate-All-Around Silicon Nanosheet MOSFETs

Chandni Akbar et al.

Summary: The study introduces an algorithm using machine learning to predict device characteristic variations, which shows the same accuracy as traditional device simulation but with reduced computation costs, driving energy-efficient devices.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Prediction of FinFET Current-Voltage and Capacitance-Voltage Curves Using Machine Learning With Autoencoder

Kashyap Mehta et al.

Summary: This study demonstrates the possibility of predicting full transistor IV and CV curves using machines trained by TCAD generated data, capturing complex physics through manifold learning. It shows that even with limited experimental data, it is possible to predict the IV/CV of novel devices before the underlying physics is well-understood.

IEEE ELECTRON DEVICE LETTERS (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic

Artificial Neural Network-Based Compact Modeling Methodology for Advanced Transistors

Jing Wang et al.

Summary: The study evaluates the effectiveness of an artificial neural network-based compact modeling methodology for advanced field-effect transistor modeling, showing that it can accurately reproduce the characteristics of advanced FETs. Through the introduction of key techniques, the capabilities of ANN models can be enhanced, along with improved efficiency and simulation turnaround time.

IEEE TRANSACTIONS ON ELECTRON DEVICES (2021)

添加到收藏夹

Article Engineering, Electrical & Electronic