A Neuromorphic Simulation Framework for Indium‐Gallium‐Zinc‐Oxide Charge‐Trap Synaptic Transistors: From Device Modeling to System Simulation

Neuromorphic computing, especially spiking neural networks (SNNs), has emerged as a promising next‐generation architecture due to its event‐driven nature and capability to emulate the computational structure of the human brain. Among its core components, indium‐gallium‐zinc‐oxide (IGZO)‐based synaptic transistors have attracted significant attention owing to their ultra‐low leakage current and low‐temperature fabrication processes. However, most existing studies have focused on device‐level characterization, leaving a realistic gap between experimental device behaviors and system‐level evaluation in large‐scale arrays. This work proposes a comprehensive framework that bridges device fabrication, mechanism‐based modeling, and system‐level simulation. A precise SPICE model based on fabricated IGZO charge‐trap synaptic transistors is developed to perform array simulations incorporating parasitic RC loads. As a result, the feasibility of the system was evaluated through SNN simulation. This systematic integration from device to system provides a novel foundation for advancing IGZO‐based neuromorphic computing technologies.