11月22日上海交通大学卢立博士学术报告

日期:2019/11/22
讲座题目:Sensor Scheduling for Remote State Estimation with Unknown Communication Channel Statistics: A Learning-Based Approach
讲座时间:2019年11月22日,下午14:00
主讲人简介

Li Lu is currently a Ph.D. candidate in Department of Computer Science and Engineering, Shanghai Jiao Tong University. He received the B.E. degree in Computer Science and Technology from Xi’an Jiaotong Universityin 2015. He was also a visiting research student in WINLAB and Department of Electrical and Computer Engineering at Rutgers University during 2018-2019 under the sponsorship of China Scholarship Council. He has published 13 papers oninternational conferences and journals (e.g., IEEE INFOCOM, ACM UbiComp, ACM MobiHoc, IEEE ToN, IEEE TPDS, IEEE TMC, etc.). He is the recipient of the Best Runner-up Poster Award of ACM MobiCom 2019. He also received the National Scholarship for Doctoral Student twice in 2018 and 2019. His research interests include mobile and ubiquitous computing, cyber security and privacy, human-computer interaction.

报告摘要

Recent years have witnessed the surge of mobile devices in our life. Thanks to the fast and convenient data connection of mobile devices, enormous users employ the devices as the frequent storage medium of their confidential information, such as personal (e.g., ID number) and financial (e.g., CVS of credit card) information. Hence, more and more users are concerned with the privacy protection of mobile devices. The ubiquitous security and privacy problem of mobile devices can be divided into two domains, i.e., defense and attack. Existing studies about defense focuson designing user authentication system, such as fingerprint, face recognition, voiceprint, etc. But they mainly rely on the physiological characteristics, which easily suffer from replay attacks. As the other side, recent researches also reveal many potential side-channel attacks leveragingmobile devices in order to arouse the privacy concern of users. This talk will present three of our work, including a lip reading-based user authentication, an acoustic-based side-channel attack, and a finger gesture-based continuous authentication for smart home. These work employ the wireless signals (including acoustics and WiFi) to sense the human activity and explore the underlying metrics to realize user authentication and side-channel attacks. Such approaches help users to understand the potential risk when using mobile devices and propose probable solutions for privacy protection