A Data-Driven ANN-Based Model for FeCAP and FeFET: Orienting to SPICE and Circuit Design
- 2 April 2026
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Electron Device Letters
- Vol. 47 (6) , 1233-1236
Abstract
Physics-based compact models for emerging non-volatile memories (NVMs) are often limited by the complex interactions of microscopic domains and defects that are difficult to capture analytically, resulting in reduced accuracy and simulation efficiency. To address this challenge, a machine learning (ML)-based approach is proposed using artificial neural networks (ANNs) trained entirely on device measurement data, enabling a direct translation of fabrication characteristics into SPICE-compatible circuit models. The resulting models achieve high accuracy (MSE: 0.724, adjusted ${R}^{{2}}$ : 0.998), significantly outperforming physics-based baselines with an $18\times $ lower MSE for polarization and a two-order-of-magnitude precision improvement in FeFET current simulation, while accurately capturing the wake-up process. Furthermore, the model demonstrates robust out-of-distribution (OOD) extrapolation to unseen ferroelectric thicknesses and a 33.7% improvement in simulation speed. These results validate the ML-based approach as a highly efficient, SPICE-compatible solution for next-generation memory.Keywords
This publication has 17 references indexed in Scilit:
- Machine Learning Assisted Compact Modeling of Dynamic Switching and Transient Response in Ferroelectric Capacitor2025
- Understanding the Time Dependent Write and Read Performance of IGZO-channel FeFETs2024
- Can Interface Layer be Really Free for HfxZr1-x
O2 Based Ferroelectric Field-Effect Transistors With Oxide Semiconductor Channel?IEEE Electron Device Letters, 2024
- Antiferroelectric Phase Evolution in HfxZr1-xO2 Thin Film Toward High Endurance of Non-Volatile Memory DevicesIEEE Electron Device Letters, 2022
- Organic Ferroelectric Materials and Applications2022
- Understanding and modelling the PBTI reliability of thin-film IGZO transistors2021
- High-Peformance BEOL-Compatible Atomic-Layer-Deposited In2O3 Fe-FETs Enabled by Channel Length Scaling down to 7 nm: Achieving Performance Enhancement with Large Memory Window of 2.2 V, Long Retention > 10 years and High Endurance > 108 Cycles2021
- Impact of Extrinsic Variation Sources on the Device-to-Device Variation in Ferroelectric FET2020
- A ferroelectric semiconductor field-effect transistorNature Electronics, 2019
- Wake-up effects in Si-doped hafnium oxide ferroelectric thin filmsApplied Physics Letters, 2013