Exploring Machine Learning Algorithms for User Activity Inference from IoT Network Traffic

Sep 26, 2023· Kuai Xu , Yinxin Wan , Xuanli Lin , Feng Wang , Guoliang Xue · 1 min read
Summary
This study infers smart-home user activities directly from overlapping IoT network traffic patterns. Wavelet analysis separates activity-related signals from background traffic, and supervised learning classifies activities using the extracted features. Experiments with labeled activities and traffic collected in real homes demonstrate accurate activity inference.
Type
Publication
2023 IEEE 20th International Conference on Mobile Ad Hoc and Smart Systems (MASS), 366-374. IEEE
publication

Abstract

The availability of ubiquitous and heterogeneous Internet-of-Things (IoT) devices in smart homes and their interactions with users provide a unique opportunity to monitor, understand, recognize, learn, and infer user activities for safety monitoring, connected health, energy saving as well as other disruptive services. Our analysis on IoT network traffic from smart homes with a variety of IoT devices has discovered that user activities often trigger overlapping traffic waves from multiple IoT devices that are deployed near the activities. This insight leads us to adopt wavelet analysis to decompose IoT network traffic in smart homes into low, middle, and high frequency bands that distinguish IoT traffic waves triggered by user activities from background noises such as heartbeat signals between IoT devices and cloud servers. Subsequently, we extract a broad range of traffic features from these IoT traffic waves and explore supervised machine learning (ML) algorithms to classify various user activities with these features. Based on the labelled user activities and IoT network traffic data collected from real smart home environments, our experiments have demonstrated that the ML-based algorithms are able to use IoT network traffic to accurately infer various user activities in smart homes.

Best Paper Award Recipient

Internet of Things Machine Learning Activity Inference
Kuai Xu
Authors
Kuai Xu
Professor of Computer Science
Kuai Xu is a professor of computer science in the School of Mathematical and Natural Sciences at Arizona State University. His research covers network security, network measurement and analysis, cloud computing, home networks, and online social networks. He received his PhD in Computer Science from the University of Minnesota in 2006 and his bachelor’s and master’s degrees in Computer Science from Peking University in 1998 and 2001, respectively.
Yinxin Wan
Authors
Yinxin Wan
Assistant Professor of Computer Science
Yinxin Wan is an assistant professor in the Department of Computer Science at the University of Massachusetts Boston. His research focuses on cybersecurity, secure and trustworthy artificial intelligence, network measurement, the Internet of Things, and quantum networking. He received his PhD in Computer Science from Arizona State University in 2023, advised by Guoliang Xue and Kuai Xu, and his bachelor’s degree in Information Security from the University of Science and Technology of China.
Xuanli Lin
Authors
Xuanli Lin
PhD Student in Computer Science

Xuanli Lin is a fifth-year PhD student in the Computer Science department at Arizona State University, supervised by Dr. Guoliang Xue.

His research interests include network optimization, network security, artificial intelligence, and the Internet of Things.

Feng Wang
Authors
Feng Wang
Professor of Applied Computing
Feng Wang is a professor in the applied computing program in the School of Mathematical and Natural Sciences at Arizona State University. Her research spans network science, social media analysis, network optimization, network security, and wireless sensor networks. She received her PhD in Computer Science from the University of Minnesota, Twin Cities, in 2005, her master’s degree in Computer Science from Peking University in 1999, and her bachelor’s degree in Computer Science from Wuhan University in 1996. She joined Arizona State University in 2007.
Guoliang Xue
Authors
Guoliang Xue
Professor of Computer Science and Engineering
Guoliang Xue is a professor in Arizona State University’s School of Computing and Augmented Intelligence and an IEEE Fellow. He investigates wireless and quantum networks, network security and privacy, and optimization. He earned his PhD in computer science from the University of Minnesota in 1991. His honors include the IEEE Communications Society’s 2019 William R. Bennett Prize, and he chaired the IEEE INFOCOM Steering Committee from 2020 through 2025.