• 文献检索
  • 文档翻译
  • 深度研究
  • 学术资讯
  • Suppr Zotero 插件Zotero 插件
  • 邀请有礼
  • 套餐&价格
  • 历史记录
应用&插件
Suppr Zotero 插件Zotero 插件浏览器插件Mac 客户端Windows 客户端微信小程序
定价
高级版会员购买积分包购买API积分包
服务
文献检索文档翻译深度研究API 文档MCP 服务
关于我们
关于 Suppr公司介绍联系我们用户协议隐私条款
关注我们

Suppr 超能文献

核心技术专利:CN118964589B侵权必究
粤ICP备2023148730 号-1Suppr @ 2026

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验

一种具有整体集成 3D 忆阻器交叉点/CMOS 混合电路的乘加引擎。

A multiply-add engine with monolithically integrated 3D memristor crossbar/CMOS hybrid circuit.

机构信息

Electrical and Computer Engineering Department, University of California, Santa Barbara, CA, 93106, USA.

Materials Department, University of California, Santa Barbara, CA, 93106, USA.

出版信息

Sci Rep. 2017 Feb 14;7:42429. doi: 10.1038/srep42429.

DOI:10.1038/srep42429
PMID:28195239
原文链接:https://pmc.ncbi.nlm.nih.gov/articles/PMC5307953/
Abstract

Silicon (Si) based complementary metal-oxide semiconductor (CMOS) technology has been the driving force of the information-technology revolution. However, scaling of CMOS technology as per Moore's law has reached a serious bottleneck. Among the emerging technologies memristive devices can be promising for both memory as well as computing applications. Hybrid CMOS/memristor circuits with CMOL (CMOS + "Molecular") architecture have been proposed to combine the extremely high density of the memristive devices with the robustness of CMOS technology, leading to terabit-scale memory and extremely efficient computing paradigm. In this work, we demonstrate a hybrid 3D CMOL circuit with 2 layers of memristive crossbars monolithically integrated on a pre-fabricated CMOS substrate. The integrated crossbars can be fully operated through the underlying CMOS circuitry. The memristive devices in both layers exhibit analog switching behavior with controlled tunability and stable multi-level operation. We perform dot-product operations with the 2D and 3D memristive crossbars to demonstrate the applicability of such 3D CMOL hybrid circuits as a multiply-add engine. To the best of our knowledge this is the first demonstration of a functional 3D CMOL hybrid circuit.

摘要

基于硅的互补金属氧化物半导体 (CMOS) 技术一直是信息技术革命的推动力。然而,按照摩尔定律对 CMOS 技术进行缩放已经达到了严重的瓶颈。在新兴技术中,忆阻器器件在存储和计算应用方面都具有广阔的前景。具有 CMOL(CMOS + “分子”)架构的 CMOS/忆阻器混合电路已经被提出,以将忆阻器器件的极高密度与 CMOS 技术的鲁棒性结合起来,从而实现太位级的内存和极其高效的计算范例。在这项工作中,我们展示了一个具有两层忆阻器交叉结构的混合 3D CMOL 电路,这些交叉结构被完全集成在预制的 CMOS 衬底上。底层的 CMOS 电路可以完全控制集成交叉结构的操作。两个忆阻器层中的器件都表现出具有可控可调谐性和稳定多级操作的模拟开关行为。我们通过二维和三维忆阻器交叉结构执行点积运算,以证明这种 3D CMOL 混合电路作为乘法-累加引擎的适用性。据我们所知,这是第一个功能 3D CMOL 混合电路的演示。

相似文献

1
A multiply-add engine with monolithically integrated 3D memristor crossbar/CMOS hybrid circuit.一种具有整体集成 3D 忆阻器交叉点/CMOS 混合电路的乘加引擎。
Sci Rep. 2017 Feb 14;7:42429. doi: 10.1038/srep42429.
2
3D integration of planar crossbar memristive devices with CMOS substrate.平面交叉阵列忆阻器件与CMOS衬底的3D集成。
Nanotechnology. 2014 Oct 10;25(40):405202. doi: 10.1088/0957-4484/25/40/405202.
3
Thousands of conductance levels in memristors integrated on CMOS.在 CMOS 上集成的数千个电导水平的忆阻器。
Nature. 2023 Mar;615(7954):823-829. doi: 10.1038/s41586-023-05759-5. Epub 2023 Mar 29.
4
Corrigendum: A multiply-add engine with monolithically integrated 3D memristor crossbar/CMOS hybrid circuit.勘误:具有单片集成3D忆阻器交叉开关/CMOS混合电路的乘法累加引擎。
Sci Rep. 2017 Jul 27;7:46874. doi: 10.1038/srep46874.
5
Four-dimensional address topology for circuits with stacked multilayer crossbar arrays.堆叠多层交叉阵列电路的四维地址拓扑结构。
Proc Natl Acad Sci U S A. 2009 Dec 1;106(48):20155-8. doi: 10.1073/pnas.0906949106. Epub 2009 Nov 16.
6
Training and operation of an integrated neuromorphic network based on metal-oxide memristors.基于金属氧化物忆阻器的集成神经形态网络的训练和操作。
Nature. 2015 May 7;521(7550):61-4. doi: 10.1038/nature14441.
7
3D CMOS/molecular hybrid circuits.
J Nanosci Nanotechnol. 2009 Feb;9(2):1015-8. doi: 10.1166/jnn.2009.c076.
8
From Memristive Materials to Neural Networks.从忆阻材料到神经网络。
ACS Appl Mater Interfaces. 2020 Dec 9;12(49):54243-54265. doi: 10.1021/acsami.0c10796. Epub 2020 Nov 24.
9
High precision tuning of state for memristive devices by adaptable variation-tolerant algorithm.通过自适应变容容忍算法对忆阻器器件进行高精度调谐。
Nanotechnology. 2012 Feb 24;23(7):075201. doi: 10.1088/0957-4484/23/7/075201. Epub 2012 Jan 20.
10
In-Memory-Computing Realization with a Photodiode/Memristor Based Vision Sensor.基于光电二极管/忆阻器的视觉传感器实现内存计算
Materials (Basel). 2021 Sep 10;14(18):5223. doi: 10.3390/ma14185223.

引用本文的文献

1
Research progress in architecture and application of RRAM with computing-in-memory.具有内存计算功能的阻变随机存取存储器的架构与应用研究进展
Nanoscale Adv. 2023 Feb 27;5(6):1559-1573. doi: 10.1039/d3na00025g. eCollection 2023 Mar 14.
2
Temporal State Machines: Using Temporal Memory to Stitch Time-based Graph Computations.时态状态机:利用时态记忆拼接基于时间的图计算
ACM J Emerg Technol Comput Syst. 2021 Jul;17(3). doi: 10.1145/3451214.
3
All-Printed Flexible Memristor with Metal-Non-Metal-Doped TiO Nanoparticle Thin Films.具有金属 - 非金属掺杂二氧化钛纳米颗粒薄膜的全印刷柔性忆阻器

本文引用的文献

1
Memristive crypto primitive for building highly secure physical unclonable functions.用于构建高度安全的物理不可克隆功能的忆阻密码原语。
Sci Rep. 2015 Aug 4;5:12785. doi: 10.1038/srep12785.
2
Training and operation of an integrated neuromorphic network based on metal-oxide memristors.基于金属氧化物忆阻器的集成神经形态网络的训练和操作。
Nature. 2015 May 7;521(7550):61-4. doi: 10.1038/nature14441.
3
3D integration of planar crossbar memristive devices with CMOS substrate.平面交叉阵列忆阻器件与CMOS衬底的3D集成。
Nanomaterials (Basel). 2022 Jul 3;12(13):2289. doi: 10.3390/nano12132289.
4
Sparse CNT networks with implanted AgAu nanoparticles: A novel memristor with short-term memory bordering between diffusive and bipolar switching.具有植入 AgAu 纳米颗粒的稀疏 CNT 网络:一种具有介于扩散和双极开关之间的短期记忆边界的新型忆阻器。
PLoS One. 2022 Mar 31;17(3):e0264846. doi: 10.1371/journal.pone.0264846. eCollection 2022.
5
A Quantized Convolutional Neural Network Implemented With Memristor for Image Denoising and Recognition.一种基于忆阻器实现的用于图像去噪与识别的量化卷积神经网络。
Front Neurosci. 2021 Sep 16;15:717222. doi: 10.3389/fnins.2021.717222. eCollection 2021.
6
Memristor-CMOS Hybrid Neuron Circuit with Nonideal-Effect Correction Related to Parasitic Resistance for Binary-Memristor-Crossbar Neural Networks.用于二元忆阻器交叉开关神经网络的、具有与寄生电阻相关的非理想效应校正的忆阻器-互补金属氧化物半导体混合神经元电路。
Micromachines (Basel). 2021 Jul 1;12(7):791. doi: 10.3390/mi12070791.
7
Ratio-based multi-level resistive memory cells.基于比率的多级电阻式存储单元。
Sci Rep. 2021 Jan 14;11(1):1351. doi: 10.1038/s41598-020-80121-7.
8
Roadmap on emerging hardware and technology for machine learning.机器学习新兴硬件与技术路线图
Nanotechnology. 2021 Jan 1;32(1):012002. doi: 10.1088/1361-6528/aba70f.
9
Improving the Recognition Accuracy of Memristive Neural Networks via Homogenized Analog Type Conductance Quantization.通过均匀模拟型电导量化提高忆阻神经网络的识别准确率
Micromachines (Basel). 2020 Apr 18;11(4):427. doi: 10.3390/mi11040427.
10
Time and rate dependent synaptic learning in neuro-mimicking resistive memories.神经模拟电阻式存储器中的时变和率相关突触学习。
Sci Rep. 2019 Oct 28;9(1):15404. doi: 10.1038/s41598-019-51700-0.
Nanotechnology. 2014 Oct 10;25(40):405202. doi: 10.1088/0957-4484/25/40/405202.
4
Stochastic memristive devices for computing and neuromorphic applications.用于计算和神经形态应用的随机忆阻器件。
Nanoscale. 2013 Jul 7;5(13):5872-8. doi: 10.1039/c3nr01176c. Epub 2013 May 22.
5
Memristive devices for computing.忆阻器计算设备。
Nat Nanotechnol. 2013 Jan;8(1):13-24. doi: 10.1038/nnano.2012.240.
6
A functional hybrid memristor crossbar-array/CMOS system for data storage and neuromorphic applications.用于数据存储和神经形态应用的功能混合忆阻器交叉阵列/CMOS 系统。
Nano Lett. 2012 Jan 11;12(1):389-95. doi: 10.1021/nl203687n. Epub 2011 Dec 9.
7
Nanoscale memristor device as synapse in neuromorphic systems.纳米级忆阻器器件作为神经形态系统中的突触。
Nano Lett. 2010 Apr 14;10(4):1297-301. doi: 10.1021/nl904092h.
8
Four-dimensional address topology for circuits with stacked multilayer crossbar arrays.堆叠多层交叉阵列电路的四维地址拓扑结构。
Proc Natl Acad Sci U S A. 2009 Dec 1;106(48):20155-8. doi: 10.1073/pnas.0906949106. Epub 2009 Nov 16.
9
Memristor-CMOS hybrid integrated circuits for reconfigurable logic.用于可重构逻辑的忆阻器-互补金属氧化物半导体混合集成电路。
Nano Lett. 2009 Oct;9(10):3640-5. doi: 10.1021/nl901874j.