SaTC: CORE: Small: Collaborative: Trustworthy Hierarchical Edge Computing

边缘计算已迅速崛起为一种有效的范式，用于捕获、处理、洞察海量物联网(IoT)数据，并在其生成地点附近采取行动。未来的边缘计算系统将是层次化和异构化的。单个物联网应用程序利用多个实体拥有的边缘计算资源将变得越来越常见。该项目设想了层次化边缘计算(HEC)服务提供商的出现，这些服务提供商从异构的多所有者边缘计算系统购买计算服务，并向单个物联网应用提供统一的边缘计算服务。港灯系统的发展将对交通、医疗、能源、教育、社会生活、公共安全等诸多领域产生深远影响。安全和隐私是阻碍HEC模式广泛发展和部署的最具挑战性的障碍之一。该项目旨在解决HEC系统中的关键安全和隐私挑战，在该挑战下，异质多所有者边缘计算系统可以通过单一服务提供商共同为最终用户提供可信计算服务。本文的研究内容包括四个方面：(1)开发具有新鲜度保证的数据流认证新技术；(2)研究基于相关随机响应的局部差异隐私数据分析技术；(3)设计一种新的分布式隐私保护协作学习框架；(4)构建一个原型HEC系统，对所提出的技术进行全面的验证和评估。如果成功，这项研究将成为边缘计算服务和物联网应用爆炸性开发和部署的关键推动因素。这项研究还将丰富网络和分布式系统安全、数据隐私和边缘计算的科学知识。大量的项目交付成果将通过教程、讲座、出版物和软件工具包在网上公开提供。此外，该项目将把研究活动与课程开发相结合，为女性和代表性不足的学生提供研究机会，通过高级设计项目增强本科生的研究经验，并通过外展计划培养K-12学生对科学、技术、工程和数学(STEM)的兴趣。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

Edge computing has quickly risen as an effective paradigm to capture, process, gain insights from, and act upon the massive amount of Internet of Things (IoT) data close to where it is generated. Future edge computing systems are expected to be hierarchical and heterogeneous. It will become increasingly common for a single IoT application to utilize edge computing resources owned by multiple entities. This project envisions the emergence of hierarchical edge computing (HEC) service providers that purchase computing services from heterogeneous multi-owner edge computing systems and provide unified edge computing services to individual IoT applications. The development of HEC systems will have a profound impact on transportation, healthcare, energy, education, social life, public safety, and many other sectors. Security and privacy are among the most challenging obstacles that hinder the wide development and deployment of the promising HEC paradigm.This project aims to tackle key security and privacy challenges in HEC systems, under which heterogeneous multi-owner edge computing systems can jointly provide trustworthy computing services to end-users via a single service provider. There are four research thrusts: (1) developing novel techniques to authenticate data streams with freshness guarantees; (2) investigating locally differentially private data analysis techniques via correlated randomized responses; (3) designing a novel framework for distributed privacy-preserving collaborative learning; and (4) building a prototype HEC system to thoroughly validate and evaluate the proposed techniques. If successful, the research can serve as a key enabler for the explosive development and deployment of edge computing services and IoT applications. The research will also enrich the scientific knowledge of network and distributed system security, data privacy, and edge computing. A substantial quantity of project deliverables will be made publicly available online through tutorials, talks, publications, and software toolkits. In addition, this project will integrate the research activities with curriculum development, provide research opportunities to female and underrepresented students, enhance undergraduate research experience through senior design projects, and foster the interest of K-12 students in science, technology, engineering, and math (STEM) via outreach programs.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.