ECRAM as Scalable Synaptic Cell for High-Speed, Low-Power Neuromorphic Computing
ECRAM as Scalable Synaptic Cell for High-Speed, Low-Power Neuromorphic Computing
复制标题
DOI：
10.1109/iedm.2018.8614551
复制
发表时间：
2018-12
期刊：
影响因子：
--
通讯作者：
Jianshi Tang;Douglas M. Bishop;Seyoung Kim;M. Copel;T. Gokmen;T. Todorov;SangHoon Shin;Ko-Tao Lee;P. Solomon;Kevin K. H. Chan;W. Haensch;J. Rozen
中科院分区：
文献类型：
--
作者：
Jianshi Tang;Douglas M. Bishop;Seyoung Kim;M. Copel;T. Gokmen;T. Todorov;SangHoon Shin;Ko-Tao Lee;P. Solomon;Kevin K. H. Chan;W. Haensch;J. Rozen
我们展示了一个非易失性电化学随机存取存储器（ECRAM）的基础上，锂（Li）离子嵌入氧化钨（WO 3）的高速，低功耗的神经形态计算。对称和线性更新的沟道电导实现使用栅极电流脉冲，其中高达1000个离散状态，具有大的动态范围和良好的保持力被证明。基于实验数据的MNIST模拟表明，准确率为96%。第一次，高速编程与脉冲宽度下降到5 ns和设备操作的规模下降到300\ \text{nm}^{2}$，证实了技术相关的ECRAM的神经形态阵列的实施。实验还证实了电导的变化与脉冲宽度、幅度和电荷成线性关系，对于100 × 100 nm ^{2}$器件，其开关能量为100 fJ。
We demonstrate a nonvolatile Electro-Chemical Random-Access Memory (ECRAM) based on lithium (Li) ion intercalation in tungsten oxide (WO3) for high-speed, low-power neuromorphic computing. Symmetric and linear update on the channel conductance is achieved using gate current pulses, where up to 1000 discrete states with large dynamic range and good retention are demonstrated. MNIST simulation based on the experimental data shows an accuracy of 96%. For the first time, high-speed programming with pulse width down to 5 ns and device operation at scales down to $300\times 300\ \text{nm}^{2}$ are shown, confirming the technological relevance of ECRAM for neuromorphic array implementation. It is also verified that the conductance change scales linearly with pulse width, amplitude and charge, projecting an ultralow switching energy ∼1 fJ for $100\times 100\ \text{nm}^{2}$ devices.