Machine learning based device modeling and performance optimization for dual gate FinFET Prashanth Kumar, R Kiran Kumar, Vinod A, Ravi Tiwari, V Shalini, et al. Engineering Research Express, 2026 In this research, we present a unique machine learning (ML)-based pipeline for simulating the I – V (current–voltage) characteristics of dual-gate fin field-effect transistor (FinFET) devices and improving them. The creative method builds models to effectively and precisely predict crucial outputs while maintaining the most reliable ML model. Several key input elements used in this work, including work function, fin height, temperature, and doping concentration, define the FinFET structure form and related doping profile. Another way we depict this is by splitting the notion of several variables and then the interactions between them in our dataset model. The ML models are then trained using this data to forecast the I – V behavior of FinFET devices. The current–voltage response is correctly predicted by the training model and has a strong correlation with traditional technology computer aided design (TCAD) simulations. Device model: less reliant on TCAD usage and time, more predictable. Fast and accurate DC characteristic forecasts are made possible by well-trained ML, which makes it a very dependable maintenance tool. The findings reveal that ML might be used to model the design and performance of advanced FinFET technologies and show that our framework achieves a strong optimization in the process with maintained high accuracy.
Temperature-Induced Changes in Multifin-Schottky Barrier FinFETs: An Analog/RF Linearity Investigation V Shalini, Prashanth Kumar Advanced Theory and Simulations, 2025 In this script, a Gallium Nitride (GaN)‐based FinFET structure is proposed with a multi‐channel device that is designed and simulated. Here, the 3D‐Sentaures TCAD simulator is used to investigate the analog/radio frequency performance and linearity of the MultiFin‐Schottky Barrier FinFET with different temperatures of 100–400 K. The proposed device underwent a temperature analysis, where critical parameters include drain current, ION/IOFF ratio, Transconductance (gm), higher‐order terms (gm2 and gm3), Gain Bandwidth Product (GBP), Cut‐off Frequency (fT), Transit Time (τ), Transconductance Generation Factor (TGF), Transconductance Frequency Product (TFP), Voltage Input Intercept Point (VIP2, VIP3), Input Intercept Point (IIP3), and Third Order Intermodulation Distortion (IMD3) is thoroughly examined. Thus, the proposed GaN‐based FinFET validates as a strong potential contender for GaN‐based analog/RF applications.
Study on recent trends of smart wearable N Anusha, B Prashanth Kumar, Lucky Agarwal 2022 2nd International Conference on Artificial Intelligence and Signal Processing Aisp 2022, 2022
