CAREER: Elastic Intermittent Computation Enabling Batteryless Edge Intelligence

职业：弹性间歇计算实现无电池边缘智能

• 批准号: 2339193
• 负责人: Arman Roohi
• 金额: $ 67.98万
• 依托单位: University of Nebraska-Lincoln
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2024
• 资助国家: 美国
• 起止时间: 2024-08-01 至 2029-07-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2339193&HistoricalAwards=false
• 关键词: CAREER; Elastic; Intermittent; Computation; Enabling

The Internet of Things (IoT) promises to advance smart homes, cities, industries, and healthcare. It is projected to reach a market value of $4.5 trillion by 2035. However, challenges include the environmental impact and disposal issues of battery-powered devices, and their limited processing power and memory. This highlights a need for sustainable, efficient IoT solutions. This CAREER project innovates duty-cycle-variable computing, proposing Elastic Intermittent Computation foundations, which will be advanced through application, architecture, and circuit-level design flows, by: (1) understanding and modeling the intermittent behavior of batteryless edge intelligence, and designing efficient intermittent-robust IP cores to ensure uninterrupted operation via a novel automated framework; and (2) exploring existing machine learning training algorithms and developing hardware-software co-design strategies for energy/intermittent-aware scheduling and execution. The technologies developed through this CAREER project promise to reduce electronic waste and carbon emissions as the world moves toward the era of a trillion connected, batteryless things. This research aims to advance sustainable computing, supporting the shift towards a carbon-neutral society. Developing batteryless processing and learning systems will enhance innovations in key sectors such as agriculture, healthcare, and defense, focusing on affordability and reliability. Alongside the research on sustainable smart devices, this CAREER project will launch hands-on education programs, from high school through the graduate level, exploring real-world applications of the underlying concepts. A Practice for Practical Problems initiative will also be introduced, emphasizing computational thinking and real-world challenges, assisting in bridging the gap between high-tech industries and conventional teaching programs. It seeks to prepare a diverse future workforce in computer science and engineering, with programs designed for high-school teachers and students. This initiative will particularly benefit rural students with limited access to such opportunities, thus broadening their educational and career horizons in technology and engineering.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.

物联网（IoT）有望推动智能家居、城市、工业和医疗保健的发展。预计到2035年，其市场价值将达到4.5万亿美元。然而，挑战包括电池供电设备的环境影响和处置问题，以及它们有限的处理能力和内存。这凸显了对可持续、高效物联网解决方案的需求。该CAREER项目创新了占空比可变计算，提出了弹性间歇计算基础，将通过应用程序，架构和电路级设计流程来推进：（1）理解和建模无电池边缘智能的间歇行为，并设计高效的容错-鲁棒IP核，以确保通过新型自动化框架实现不间断运行;以及（2）探索现有的机器学习训练算法，并开发用于能量/不确定性感知调度和执行的硬件-软件协同设计策略。通过这个CAREER项目开发的技术有望减少电子废物和碳排放，因为世界正在走向一万亿个连接的无电池时代。这项研究旨在推进可持续计算，支持向碳中和社会的转变。开发无电池处理和学习系统将加强农业、医疗保健和国防等关键领域的创新，重点是可负担性和可靠性。除了对可持续智能设备的研究外，该CAREER项目还将推出从高中到研究生阶段的实践教育计划，探索基本概念的现实应用。还将引入实践问题倡议，强调计算思维和现实世界的挑战，协助弥合高科技产业和传统教学计划之间的差距。它旨在为计算机科学和工程领域的多元化未来劳动力做好准备，为高中教师和学生设计课程。这一举措将使机会有限的农村学生特别受益，从而拓宽他们在技术和工程方面的教育和职业视野。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。