神经形态工程学
记忆电阻器
材料科学
长时程增强
双层
光电子学
电导
神经促进
突触重量
突触可塑性
异质结
纳米技术
计算机科学
电子工程
化学
人工智能
人工神经网络
膜
物理
凝聚态物理
生物化学
受体
工程类
作者
Sobia Ali Khan,Mehr Khalid Rahmani,Muhammad Umair Khan,Jungmin Kim,Jinho Bae,Moon Hee Kang
标识
DOI:10.1149/1945-7111/ac7754
摘要
The progress of artificial synaptic devices is vital to the realization of the consciousness of effectual bioinspired neuromorphic computing systems. In this study, we fabricated an ITO/ZnO/PTAA/Ag artificial synaptic memristor based on a low-cost solution process. The fabricated device exhibited uniform gradual bipolar resistive switching with excellent endurance and self-rectifying behavior owing to the bilayer heterojunction structure of ZnO/PTAA. The growth of the conducting filament can be efficiently controlled by modulating the current compliance and voltage during the SET process. Modification of conductance states was also observed by simulations to stimuli, which are essential for synaptic function in neuromorphic computing. Various pulse measurements were performed to mimic synaptic behaviors, including long-term potentiation, long-term depression, spike-rate-dependent plasticity, paired-pulse facilitation, and post-tetanic potentiation. Moreover, we reveal that the real device shows an approximately similar pattern recognition rate as the ideal device owing to a more uniform conductance update.
科研通智能强力驱动
Strongly Powered by AbleSci AI