德拉姆
电阻随机存取存储器
材料科学
非易失性随机存取存储器
光电子学
联轴节(管道)
电子工程
半导体存储器
电气工程
内存刷新
计算机存储器
工程类
电压
冶金
作者
Yanbo Su,Mingcheng Shi,Jianshi Tang,Yijun Li,Yiwei Du,Ran An,Jiaming Li,Yuankun Li,Jian Yao,Ruofei Hu,Yuan He,Yue Xi,Qingwen Li,Song Qiu,Qingtian Zhang,Liyang Pan,Bin Gao,He Qian,Huaqiang Wu
标识
DOI:10.1109/ted.2024.3372937
摘要
Computing-in-memory (CIM) based on analog resistive random access memory (RRAM) emerges as an energy-efficient technology for edge artificial intelligence (AI), where a large amount of ON-chip data buffer is needed to implement complex neural networks. In this work, we report a novel InGaZnOx (IGZO)/carbon nanotube (CNT) hybrid-polarity 2T0C DRAM as a backend-of-the-line (BEOL) compatible buffer, which is a monolithic 3-D (M3D) integrated with HfO2-based analog RRAM array and Si CMOS logic to demonstrate a M3D-BRIC chip. The structural integrity and proper function of each layer are systematically verified. In particular, by incorporating n-type ultralow leakage IGZO field-effect transistor (FET) for write transistor and p-type high-current CNT-FET for read, this unique hybrid-polarity 2T0C design achieves a decent retention and desirably large read currents. It also helps enhance the effective sensing window and, more importantly, resolve the charge injection issue via counteractive coupling. To demonstrate the computational advantage of M3D-BRIC architecture, a typical high-resolution (Hi-Res) video processing task is further implemented using the YOLOv3 network for object detection. The benchmark shows that the M3D-BRIC chip with BEOL 2T0C DRAM could achieve a $48.25\times $ higher processing capability compared to its 2-D counterpart.
科研通智能强力驱动
Strongly Powered by AbleSci AI