Privacy against Statistical Matching
Smart cities, connected vehicles, smart homes, and connected healthcare devices are examples of how the Internet of Things (IoT) will revolutionize our lives in the decades ahead. However, the potential loss of privacy is a significant threat to IoT penetration. This leads to a great challenge: IoT devices will be generating an astounding amount of data every second in the near future, and, even if privacy-protecting mechanisms are employed, significant privacy leaks can occur due to the sheer amount of the data generated and powerful statistical inference techniques available to the potential adversaries.
In this talk, we discuss the theoretical limits of IoT privacy. The main idea is that a large class of IoT privacy problems can be modeled mathematically as matching and de-noising time-series data. Using this setting, we introduce the information theoretic notion of perfect privacy. We consider a network of a large number of users, and assume anonymization and/or obfuscation techniques are used to preserve the privacy of users. We provide conditions under which perfect privacy is achievable. Also, we provide converse results where an adversary is able to successfully recover the data. Two models for users’ data are considered: i.i.d. as well as Markov chain-based models. We also consider the impact of correlation between user traces on their privacy. Specifically, we show that this correlation leaks a significant amount of information and that independent obfuscation of the data traces is often insufficient to remedy such. Finally, we discuss how the users can employ dependency in their obfuscation to improve their privacy.
Hossein Pishro-Nik is an Associate Professor of electrical and computer engineering at the University of Massachusetts, Amherst. He received a B.S. degree from Sharif University of Technology, and M.Sc. and Ph.D. degrees from the Georgia Institute of Technology, all in electrical and computer engineering. His research interests include information and coding theory, stochastic analysis of wireless networks, and vehicular communications. He has received an NSF Faculty Early Career Development (CAREER) Award, an Outstanding Junior Faculty Award from UMass, and an Outstanding Graduate Research Award from the Georgia Institute of Technology.