Resistive random-access memory (ReRAM) memristors are promising candidates for various compute in memory and flow-based computing approaches. As an alternative to traditional von Neumann computation, flow-based computing avoids serial movement of data between memory and processor. In this paper, we demonstrate arrays of 1 transistor 1 ReRAM (1T1R) to detect edges between 8 bit pixels using flow-based computing, and the effects of stochastic variation of ReRAM on edge detection outputs. Three different tRoff/Ron resistance ratios (1.5:1, 2.5:1 or 28.6:1) were utilized to implement multiple flow-based edge detection computation matrices for 8 bit pixels. Edge detection was distinguishable for all Roff/Ron ratios used, for all flow-based computing matrices. However, the binary output resistance ratio of the matrices improved 3-fold when the patterned Roff/Ron ratio was increased to 28.6:1. A Gaussian simulation of ReRAM resistance variability validates the experimental data, with a correlation coefficient (r) of 0.9547. These results suggest a trade-off between the flow-based edge detection output ratio and the variability of the ReRAM resistance in Roff/Ron resistance ratio.

中文翻译：

---

使用基于 HfO2 的 ReRAM 阵列研究基于流的边缘检测计算的 ReRAM 可变性


电阻式随机存取存储器 (ReRAM) 忆阻器是各种内存计算和基于流的计算方法的有希望的候选者。作为传统冯诺依曼计算的替代方案，基于流的计算避免了内存和处理器之间的数据串行移动。在本文中，我们演示了 1 个晶体管 1 个 ReRAM (1T1R) 阵列，使用基于流的计算来检测 8 位像素之间的边缘，以及 ReRAM 随机变化对边缘检测输出的影响。利用三种不同的 tRoff/Ron 电阻比（1.5:1、2.5:1 或 28.6:1）来实现 8 位像素的多个基于流的边缘检测计算矩阵。对于所有使用的 Roff/Ron 比率、所有基于流的计算矩阵，边缘检测都是可区分的。然而，当图案化的 Roff/Ron 比率增加到 28.6:1 时，矩阵的二元输出电阻比提高了 3 倍。 ReRAM 电阻变化的高斯模拟验证了实验数据，相关系数 (r) 为 0.9547。这些结果表明基于流的边缘检测输出比和 Roff/Ron 电阻比中 ReRAM 电阻的可变性之间的权衡。