Mining IoT Network Traffic in Smart Homes: Traffic Measurement, Pattern Recognition, and Security Applications
Add to Google Calendar
Date: Tue, March 21, 2023
Time: 10:30am - 11:30am
Location: Holmes Hall 244
Speaker: Yinxin Wang, Arizona State University
Date: Tue, March 21, 2023
Time: 10:30am - 11:30am
Location: Holmes Hall 244
Speaker: Yinxin Wang, Arizona State University
Abstract
Recent advances in cyber-physical systems, artificial intelligence, and cloud computing have driven the widespread deployment of IoT devices in smart homes. However, the spate of cyber-attacks exploiting the vulnerabilities and insufficient security management of smart home IoT devices have highlighted the urgency and challenges of designing efficient mechanisms for detecting, analyzing, and addressing security threats towards them. In this talk, I will present our recent efforts in securing smart home networks from the perspectives of traffic measurement, pattern recognition, and security applications. (1) I will first demonstrate our multidimensional smart home network traffic measurement framework, which has enabled us to deeply understand the smart home IoT ecosystem and detect various vulnerabilities in smart homes. (2) We further identified that critical security-related patterns, including IoT device events and user activities, can be extracted from the encrypted smart home network packets. (3) Based on our extensive knowledge of smart home systems, we proposed and implemented different schemes for securing smart home networks including abnormal network traffic detection across multiple IoT networking protocol layers, smart home safety monitoring, and system-level IoT vulnerability analysis.