CRII: SaTC: Towards Understanding Typing Privacy: Vulnerabilities and Protection

University Of Oklahoma Norman Campus, Norman OK

Investigators

Abstract

Typing is ubiquitous and plays a significantly important role in the interaction between humans and computer devices. Emerging work has identified a class of attacks that can eavesdrop keystrokes in a non-invasive way without pre-infecting the target computer with malware. Their common underlying principle is that pressing a key on a keyboard causes subtle environmental impacts unique to that key, which can be observed and correlated for all keys. However, those techniques require a training phase to establish the relationship between an observed environmental change and the action of pressing a specific key. Such a training phase is impractical for most attack scenarios. This project identifies a new type of non-invasive keystroke inference technique without requiring the training phase. The goal of this project is to systematically exploit the side effects associated with the typing process via a keyboard, understand new security threats, and design effective defense mechanisms against such attacks. This research will develop a novel type of typing privacy inference technique targeting various input such as letters and numbers in real-world scenarios and build corresponding defenses to protect typing privacy against those keystroke eavesdropping attacks. The project will provide a much-needed insight into the security risks of typing privacy disclosure, and also substantially improve the security of various computer systems with keyboard interfaces. The results from this research will be openly shared. This project will also offer graduate and undergraduate students opportunities to gain research experience in security and privacy areas. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.