报告人
Rui Xiao Zhejiang University 时间 2025年3月4日 星期二 上午 10:00-11:00 地点
308会议室
Abstract The Artificial Internet-of-Things (AIoT) has shown immense potential in driving pervasive intelligence, represented by innovations ranging from smart homes to autonomous vehicles. However, the close interaction of AIoT devices with the physical world, referred to as physical context, significantly expands the attack surface compared to purely cyberspace systems. Attackers can exploit these new vulnerabilities to gain access to sensitive information by analyzing the unintentional physical emanation. Additionally, attackers can undermine data integrity through physical manipulation and spoofing attacks. These breaches lead to severe security consequences, including privacy leakage, operational failures, and real-world accidents, which ultimately undermine the trustworthiness of AIoT systems. In this talk, I will present my work on enabling trustworthy AIoT systems within physical context. My work covers three critical components of AIoT systems: sensing, perception, and connectivity. I will demonstrate how physical context introduces significant, real-world security threats to sensors, AI infrastructure, and wireless networks, using both theoretical cross-layer models and real-world system implementations. Additionally, I will introduce novel, physical-based defense mechanisms that complement traditional software-based approaches to enable secure, private, and reliable AIoT systems. Finally, I will discuss my future research directions aimed at building truly trustworthy AIoT systems.
Biography Rui Xiao is a final-year Ph.D. candidate at Zhejiang University. His research focuses on the system security of the Artificial Internet-of-Things (AIoT) by ensuring its trustworthy interaction with the physical world and addressing critical challenges at the intersection of security and mobile/wireless/sensing systems. He has published first-authored papers in several top mobile and security conferences, including ACM MobiCom, MobiSys, SenSys, and NDSS, and has won Best Paper Award Candidate at SenSys 2021. He received his B.S. in Physics with honors at Zhejiang University in 2020. 肖睿，浙江大学博士生，2020年于浙江大学竺可桢学院获学士学位，主要研究领域为智能物联系统安全与智能感知，具体包括针对智能物联系统的侧信道安全评估与防护，以及利用泛在物理信号实现场景普适的通信与感知，并赋能下游可信应用，曾与蚂蚁集团开展深度合作，以物理层注入为媒介从而实现蚂蚁链上信息源头可信。近年来以第一作者身份在移动计算和安全顶会（包括ACM MobiCom、MobiSys、SenSys和NDSS等）上发表论文共五篇，曾获ACM SenSys 2021最佳论文奖提名。
