A Neuromorphic Brain Interface Based on RRAM Crossbar Arrays for High Throughput Real-Time Spike Sorting
A Neuromorphic Brain Interface Based on RRAM Crossbar Arrays for High Throughput Real-Time Spike Sorting
复制标题
基于 RRAM 交叉阵列的神经形态脑接口，用于高通量实时尖峰排序
DOI：
10.1109/ted.2021.3131116
复制
发表时间：
2022
影响因子：
3.1
通讯作者：
Kuzum, Duygu
中科院分区：
文献类型：
--
作者：
Shi, Yuhan;Ananthakrishnan, Akshay;Oh, Sangheon;Liu, Xin;Hota, Gopabandhu;Cauwenberghs, Gert;Kuzum, Duygu
实时的棘波分类和处理对于闭环脑机接口和神经假体来说是至关重要的。具有数百个电极的高密度多电极阵列的最新发展使得能够同时记录大量神经元的尖峰电位。然而，高通道数对实时尖峰排序硬件(HW)在数据传输带宽和计算复杂度方面提出了严格的要求。因此，有必要开发一种专门的实时硬件，它可以在消耗最低功耗的情况下，以高吞吐量在飞行中对神经棘波进行分类。在这里，我们提出了一个实时的，低延迟的尖峰排序处理器，它利用高密度的CuOxR交叉开关以大规模并行的方式实现内存中的尖峰排序。我们开发了一种与CMOS线后端(BEOL)集成兼容的制造工艺。我们广泛地描述了CuOxMemory器件的开关特性和统计变化。为了实现基于交叉开关阵列的脉冲排序，我们提出了一种基于模板匹配的脉冲排序算法，该算法可以直接映射到阻性随机存取存储器(RRAM)交叉开关上。通过使用合成和细胞外刺激的活体记录，我们实验证明了高精度的能量高效的刺激分类。与基于现场可编程门阵列(FGA)和微控制器的其他硬件实现相比，我们的神经形态界面在面积(更小的面积)、功耗(更少的功率)和延迟(对100个通道进行排序的延迟)方面提供了显著的改进。
Real-time spike sorting and processing are crucial for closed-loop brain–machine interfaces and neural prosthetics. Recent developments in high-density multielectrode arrays with hundreds of electrodes have enabled simultaneous recordings of spikes from a large number of neurons. However, the high channel count imposes stringent demands on real-time spike sorting hardware (HW) regarding data transmission bandwidth and computation complexity. Thus, it is necessary to develop a specialized real-time HW that can sort neural spikes on the fly with high throughputs while consuming minimal power. Here, we present a real-time, low latency spike sorting processor that utilizes high-density CuOxresistive crossbars to implement in-memory spike sorting in a massively parallel manner. We developed a fabrication process that is compatible with CMOS back end of line (BEOL) integration. We extensively characterized switching characteristics and statistical variations of the CuOxmemory devices. In order to implement spike sorting with crossbar arrays, we developed a template matching-based spike sorting algorithm that can be directly mapped onto resistive random-access memory (RRAM) crossbars. By using synthetic andin vivorecordings of extracellular spikes, we experimentally demonstrated energy-efficient spike sorting with high accuracy. Our neuromorphic interface offers substantial improvements in area (less area), power (less power), and latency (latency for sorting 100 channels) for real-time spike sorting compared to other HW implementations based on field-programmable gate arrays (FPGAs) and microcontrollers.