当前位置: X-MOL 学术IEEE Electron Device Lett. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
A Single Transistor Neuron with Independently Accessed Double-Gate for Excitatory-Inhibitory Function and Tunable Firing Threshold Voltage
IEEE Electron Device Letters ( IF 4.1 ) Pub Date : 2020-08-01 , DOI: 10.1109/led.2020.3001953
Joon-Kyu Han , Myungsoo Seo , Ji-Man Yu , Yoon-Je Suh , Yang-Kyu Choi

For the first time, a single transistor neuron (1T-neuron) with an independently accessed double-gate structure is demonstrated for excitatory-inhibitory function, and control of the firing threshold voltage ( ${V}_{{\text {T,firing}}}$ ). The double-gate is composed of a primary gate and a secondary gate. The former is to determine an excitatory and inhibitory state as a toggle switch and the latter is to dynamically adjust ${V}_{\text {T,firing}}$ for homeostasis to sustain reliable neuromorphic operation. In addition to its conventional excitation function, the newly implemented inhibition function improves the energy efficiency of the neuromorphic network. The tunable threshold voltage for firing maintains homeostasis, even for high or low transmitted signals that are out of a desired spiking frequency range.

中文翻译:

具有独立访问双栅极的单晶体管神经元,用于兴奋抑制功能和可调激发阈值电压

首次展示了具有独立访问的双门结构的单个晶体管神经元(1T-神经元)的兴奋抑制功能和激发阈值电压的控制。 ${V}_{{\text {T,firing}}}$ )。双闸门由一级门和二级门组成。前者是将兴奋和抑制状态确定为拨动开关,后者是动态调整 ${V}_{\text {T,firing}}$ 用于体内平衡以维持可靠的神经形态操作。除了传统的激励函数外,新实现的抑制函数提高了神经形态网络的能量效率。用于触发的可调阈值电压保持稳态,即使对于超出所需尖峰频率范围的高或低传输信号也是如此。
更新日期:2020-08-01
down
wechat
bug