Simulation Toolchain for Neuromorphic Oscillatory Neural Networks Based on Beyond-CMOS Vanadium Dioxide Devices - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Simulation Toolchain for Neuromorphic Oscillatory Neural Networks Based on Beyond-CMOS Vanadium Dioxide Devices

Aida Todri-Sanial

Résumé

In this work, we illustrate a simulation toolchain using technology computer-aided design (TCAD) software to simulate neuro-mimicking Oscillatory Neural Networks (ONNs) based on Beyond CMOS Vanadium Dioxide (VO2) oscillators. We use a dedicated TCAD approach to simulate thermal-induced resistive switching in VO2. We perform TCAD and TCAD -SPICE mixed-mode simulations to simulate all the key elements of the ONN system: VO2 device, VO2 oscillator and dynamics of coupled VO2 oscillators. This demonstrates that TCAD multi-physics simulations are an essential tool for probing the interplay between VO2 material properties, device geometry and circuit dynamics, while providing insights and guidelines for the development of ONN technology based on Beyond CMOS VO2 oscillators.
Fichier non déposé

Dates et versions

hal-03670565 , version 1 (17-05-2022)

Identifiants

Citer

Stefania Carapezzi, Corentin Delacour, Aida Todri-Sanial. Simulation Toolchain for Neuromorphic Oscillatory Neural Networks Based on Beyond-CMOS Vanadium Dioxide Devices. FLEPS 2022 - IEEE International Conference on Flexible and Printable Sensors and Systems, Jul 2022, Wien, Austria. pp.1-4, ⟨10.1109/FLEPS53764.2022.9781525⟩. ⟨hal-03670565⟩
43 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More