南京大学计算机软件新技术国家重点实验室   

摘 要：

Recently, continuous user authentication (CA) has shown its great potential in improving the user experience by allowing smart devices to perform low-effort and timely user authentication in people's daily lives. Most existing CA systems either rely on dedicated sensors or require specific inputs (e.g., gait or routine behaviors). In this work, we devise a photoplethysmography (PPG)-based CA system, which uses the PPG sensors readily available in commodity wrist-worn wearable devices for user authentication. Compared to the existing approaches, our system does not require any users' specific inputs but only the wearable device that has already been pervasively used in our daily lives. Notably, we explore the uniqueness of the human cardiac system captured by the PPG sensing technology and design a robust motion artifacts (MA) filtering method to mitigate the impact of MA from the daily activities. Additionally, we explore the general characteristic of fiducial features from PPG measurements to efficiently distinguish human cardiac systems. Furthermore, we develop an adaptive cardiac-based classifier for CA using the gradient boosting tree (GBT). Experiments with the prototype of the wrist-worn PPG sensing platform and 20 participants in different scenarios demonstrate that our system can effectively remove MA and achieve over 90% authentication success rate.

报告人简介:

Yan Wang is an Assistant Professor with the Department of Computer Science at Binghamton University, State University of New York. He received his Ph.D. degree in Electrical and Computer Engineering from Stevens Institute of Technology. His research interests include Internet of Things, Cyber Security and Privacy, Smart Healthcare, Mobile Sensing and Computing, and Connected Vehicles. He has published one book chapter, more than 26 journal and conference papers in premium conferences and peer-reviewed journals including ACM MobiCom, ACM MobiSys, ACM MobiHoc, ACM CCS, IEEE INFOCOM, IEEE ICDCS, IEEE TMC, etc. He holds 3 US patents, two of which are under commercialization. He is the recipient of three Best Paper Awards from IEEE CNS 2018, IEEE SECON 2017 and ACM ASIACCS 2016. His research has received broad press coverage, including BBC News, Yahoo News, MIT Technology Review, NBC New York, WCBS TV and Voice of America TV, etc. His research is supported by the National Science Foundation (NSF). For more information, please refer to: http://www.cs.binghamton.edu/~yanwang/index.html

时间：4月8日   10:00-11:30

地点：计算机科学技术楼229室