Security for Pervasive Edge Computing Ecosystems

点击翻译按钮获取中文摘要

Diverse, novel applications, such as autonomous driving, Industrial Internet of Things (IIoT), Augmented and Virtual Reality (AR/VR), tele-medicine, and distributed machine learning are driving the rapid growth in research and development in edge computing deployed in the last mile of the network where dynamic and resource constrained wireless communications dominate. These applications require computation to happen close to the data source and be coordinated across nearby devices, base stations, or access points as well as with data services such as stored media download at the edge. They may also require computation offloading to a data center located near the edge or somewhere on the Internet for processes such as image annotation and machine learning. Furthermore, these applications rely on ultra-low latency communications, reliable computations, and security and privacy at the network edge. This project aims to investigate the vision of a secure Pervasive Edge Computing (PEC) environment where the abundant edge computing resources (e.g. city infrastructure, individuals’ devices, and future wireless devices), can be easily and securely marshaled, in a trusted manner, to make the participation of any device in edge computing feasible. The particular research problem addressed in this project is access control and authorization of devices and services at the edge—essential for the broad adoption of PEC to meet the needs of future applications in healthcare, transportation, smart cities, and the smart grid. This project focuses on investigating the access control and authorization mechanisms in an information-centric PEC network. The researchers will investigate mechanisms for delegating access control-as-a-service (ACaaS) to the network edge; efficient revocation of access credentials to data and services in a dynamic manner; and secure accounting of the access control operations to verify access control claims of the service and also to assess users’ usage. They will contrast their designs with competing designs in the IP network to provide comprehensive comparative assessments of the efficacy of proposed approaches. The results from this project will help increase the security, adoptability, and trust in edge computing and speed up its adoption by enabling more of the edge devices to become part of edge computing.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.