CAREER: Repurposable Devices for a Greener Internet of Things

该奖项全部或部分由2021年美国救援计划法案（公法117-2）资助。组成物联网（IoT）的传感器和执行器必须在现场进行物理部署，以实现新的应用。这些地点往往难以进入或进入成本很高，例如在整个大型农场进行精准农业或在工厂内。确保物联网设备在部署后很长一段时间内都是有用和高效的，可以降低其总体成本并提高其实用性。然而，当今的许多计算设备被设计为具有短的、固定的寿命，并且硬件进步不断地使旧设备过时。该项目提出了必要的发展，以确保部署的物联网设备在初始调试后数十年仍然可用，高效和安全。其结果将是新的自供电能量收集物联网设备，可以适应不断变化的环境，新的应用需求和更新的安全要求，而无需被删除和替换。实现这一点是几个基础性的进步，包括（i）细粒度硬件和编译器强制软件模块化，实现节能软件更新，（ii）用于长期能源预测和管理的内存高效强化学习算法，以及（iii）透明的应用程序卸载，以支持未来在已经部署的物联网设备硬件上运行的软件。最终，这将使物联网设备的使用寿命与它们所连接的基础设施相似。追求数据驱动的方法来减少能源使用和温室气体的产生对于应对气候变化至关重要。该项目将开发持久和可扩展的传感器，以提供数十年寿命的可操作数据。确保较长的使用寿命将有助于避免加剧数十亿个一次性物联网设备日益严重的电子垃圾问题。该项目的技术将被纳入研究生和本科生课程，培训了解物联网，跨学科应用和道德交叉的新工程师。代表性不足的少数民族研究生和一年级本科生将在这个项目上合作，开发新的基础科学技术，并学习技术，建筑环境和相关利益相关者之间的联系。该奖项反映了NSF的法定使命，并被认为值得通过使用基金会的知识价值和更广泛的影响审查标准进行评估来支持。

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).The sensors and actuators that comprise the Internet of Things (IoT) must be physically deployed in-situ to enable new applications. These locations are often difficult or expensive to access, such as throughout a large farm for precision agriculture or within an operating factory. Ensuring the IoT devices are useful and productive long after they are deployed reduces their overall cost and increases their utility. However, many computing devices today are designed with short, fixed lifetimes, and hardware advances continuously render old devices obsolete. This project proposes the necessary developments to ensure deployed IoT devices remain usable, efficient, and secure decades after their initial commissioning. The result will be new self-powered energy-harvesting IoT devices that can adapt to changing environments, new application requirements, and updated security requirements without being removed and replaced. Enabling this are several foundational advancements, including (i) fine-grained hardware and compiler enforced software modularity that enables energy-efficient software updates, (ii) memory-efficient reinforcement learning algorithms for long-term energy prediction and management, and (iii) transparent application offloading to support future software running on already deployed IoT device hardware. Ultimately, this will equip IoT devices with lifespans similar to the infrastructure they are attached to.Pursuing data-driven approaches to reduce energy use and greenhouse gas production is essential for combating climate change. This project will develop long-lasting and scalable sensors to provide actionable data over multi-decade lifetimes. Ensuring long operational lifetimes will help avoid exacerbating a growing e-waste problem from billions of otherwise disposable IoT devices. The techniques from this project will be incorporated into graduate and undergraduate courses, training new engineers who understand the intersection of IoT, cross-disciplinary applications, and ethics. Underrepresented minority graduate students and first-year undergraduates will collaborate on this project to develop new fundamental scientific techniques and learn the connections between technology, the built environment, and the associated stakeholders. The outputs of this work, both teaching and research, will be open sourced and disseminated broadly.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.