Systematic assessment of security, privacy and usability of virtual reality learning environment applications

Abstract

[ACCESS RESTRICTED TO THE UNIVERSITY OF MISSOURI--COLUMBIA AT REQUEST OF AUTHOR.] Social Virtual Reality Environments is a new cloud computing based platform which integrates IoT to build applications in areas such as military, education, surgical training, etc. Although VR can be used for critical applications, it is important to ensure security, privacy and usability of the applications which has not been studied in depth. In this thesis, we explore new security and privacy issues in VR and their impact on overall quality of user experience. We also perform a usability study for a social VR application and show that VR based learning environment can be more effective than a traditional desktop-based environment. For systematic assessment, we propose a novel formal methods based framework to study these applications from security, privacy and usability perspectives. Our framework uses the UPPAAL tool to convert attack trees into Network of Stochastic Timed Automata (NSTA). Next, we use statistical model checking (UPPAAL SMC) to perform vulnerability assessment of the threats. Such an analysis helps us adopt pertinent design principles such as hardening, diversity and principle of least privilege to enhance the resilience of the VR systems.