空中骑兵
神经形态工程学
铁磁性
能量(信号处理)
凝聚态物理
物理
记忆电阻器
垂直的
各向异性
磁电阻
材料科学
人工神经网络
光电子学
磁场
计算机科学
光学
人工智能
量子力学
数学
几何学
作者
Ravish Kumar Raj,Namita Bindal,Sandeep Soni,Brajesh Kumar Kaushik
摘要
The development of energy-efficient and ultrafast neuromorphic computing based on the dynamics of the ferromagnetic (FM) skyrmion on the nanotrack has attained considerable interest. In this work, FM skyrmion based artificial neuron device is proposed. The perpendicular magnetic anisotropy (PMA) gradient is created on a thin film ferromagnetic (FM) layer by voltage control-PMA effect (VC-PMA). The anisotropy is directly co-related with the strength of 𝑚𝑧 that affects the size of skyrmion meaning that in the region with larger PMA, the skyrmion size is smaller and hence, more energy. However, the skyrmions have the tendency to move in the direction to minimize the energy. Hence, the skyrmion move towards the lower PMA. This behavior of skyrmion on a nanotrack with PMA gradient corresponds to the leaky-integrate-fire (LIF) functionality of the neuron device. Hence, the suggested energy-efficient artificial neuron opens up the path for developing for energy-efficient neuromorphic computing.
科研通智能强力驱动
Strongly Powered by AbleSci AI