FuSe-TG: OpenBrain: Open-Source Cross-Layer Design of RRAM-based Neuromorphic Systems

FuSe-TG：OpenBrain：基于 RRAM 的神经形态系统的开源跨层设计

• 批准号: 2235440
• 负责人: Matthew Guthaus
• 金额: $ 30万
• 依托单位: University of California-Santa Cruz
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2023
• 资助国家: 美国
• 起止时间: 2023-04-15 至 2025-03-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2235440&HistoricalAwards=false
• 关键词: FuSe; TG; OpenBrain; Open; Source

The OpenBrain project will create a shareable framework that assists researchers in creating computer chips that can “think” in ways inspired by the human brain. This teaming grant specifically brings together researchers with promising specialties ranging from electronic devices to deep-learning applications so that they may synergize across traditional system boundaries. The team will engage the research and education communities through the accessibility of open-source so that others may harness OpenBrain, reproduce results, and iterate upon prior results.The OpenBrain project will advance computing by creating sustainable smart devices that use brain-inspired techniques for learning with improved power and performance over traditional computer processors. These new computer chips will improve how we store, manage, and harness information in sensor systems for applications in medicine, the environment, transportation, and many other areas. The open-source nature of the project will also foster education and workforce development by enabling students to have insight into and work directly with leading-edge chip technologies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

OpenBrain项目将创建一个可共享的框架，帮助研究人员创建能够以人脑启发的方式“思考”的计算机芯片。这项团队资助专门汇集了从电子设备到深度学习应用等有前途的专业研究人员，以便他们可以跨越传统系统边界进行协同。该团队将通过开放源代码的可访问性吸引研究和教育界的参与，以便其他人可以利用OpenBrain，复制结果，并在先前的结果基础上进行扩展。OpenBrain项目将通过创建可持续的智能设备来推进计算，这些设备使用大脑启发的技术进行学习，与传统的计算机处理器相比具有更高的功率和性能。这些新的计算机芯片将改善我们在传感器系统中存储、管理和利用信息的方式，以应用于医学、环境、交通和许多其他领域。该项目的开源性质还将通过使学生能够深入了解并直接使用领先的芯片技术来促进教育和劳动力发展。该奖项反映了NSF的法定使命，并通过使用基金会的智力价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

OpenSpike: An OpenRAM SNN Accelerator

OpenSpike：OpenRAM SNN 加速器

• DOI: 10.1109/iscas46773.2023.10182182
• 发表时间: 2023
• 期刊: Proceedings IEEE International Symposium on Circuits and Systems
• 作者: Modaresi, Farhad;Guthaus, Matthew;Eshraghian, Jason K.
• 通讯作者: Eshraghian, Jason K.

SRAM Design with OpenRAM in SkyWater 130nm

SkyWater 130nm 中采用 OpenRAM 的 SRAM 设计

• DOI: 10.1109/iscas46773.2023.10181379
• 发表时间: 2023
• 期刊: Proceedings IEEE International Symposium on Circuits and Systems
• 作者: Cirimelli-Low, Jesse;Khan, Muhammad Hadir;Crow, Samuel;Lonkar, Amogh;Onal, Bugra;Zonenberg, Andrew D.;Guthaus, Matthew R.
• 通讯作者: Guthaus, Matthew R.

EMBER: A 100 MHz, 0.86 mm2, Multiple-Bits-per-Cell RRAM Macro in 40 nm CMOS with Compact Peripherals and 1.0 pJ/bit Read Circuitry

EMBER：采用 40 nm CMOS 封装的 100 MHz、0.86 mm2、每单元多位 RRAM 宏，具有紧凑型外设和 1.0 pJ/位读取电路

• DOI: 10.1109/esscirc59616.2023.10268807
• 发表时间: 2023
• 期刊: ESSCIRC 2023- IEEE 49th European Solid State Circuits Conference (ESSCIRC
• 作者: Upton, Luke R.;Levy, Akash;Scott, Michael D.;Rich, Dennis;Khwa, Win-San;Chih, Yu-Der;Chang, Meng-Fan;Mitra, Subhasish;Raina, Priyanka;Murmann, Boris
• 通讯作者: Murmann, Boris