SaTC: CORE: Small: Modeling and Defense of Cyber Attacks for Improving Social Virtual Reality Resilience

点击翻译按钮获取中文摘要

Social Virtual Reality (VR) Learning Environment (VRLE) technologies offer a new medium for flexible learning environments with geo-distributed users. Secure Social VRLEs can augment human performance in a wide range of domains including special education, disaster response training, and healthcare. This project’s goal is to investigate Social VRLEs deployed on networked systems that need to be properly designed for performance and resilience to prevent unique cyber-attacks. The consequences of ill-suited design make Social VRLEs vulnerable to security breaches and privacy leakage attacks that significantly impact users (e.g., STEM education students, first responders, patients). More severely, poor design as well as security and privacy attacks can cause disruption of user immersive experience. Hence, the project activities are focused on ensuring the security, privacy, and safety (SPS) in social VRLEs to enable safe and effective student learning activities. The project outcomes include theory and techniques for defense of attacks in social VRLEs. These outcomes will be disseminated to the broader community via open-source code, educational materials, peer-reviewed publications, design case studies, and data sets to foster the setup of trustworthy learning environments with a co-operative setting of geographically distributed users and VR devices. Towards meeting the above goal, this project aims to create a transformative co-design of learning and resilient aspects of social VRLEs. Project activities help address the knowledge gap in the understanding and quantitative measurement of how cyber-attacks influence the system and network factors that hinder the uninterrupted user immersive experience. Examples of these attacks include immersion attacks such as occlusion attack, Chaperone file attack and network fault scenarios. More specifically, the project objectives are: (i) to investigate techniques for formal modeling and scalable analysis of SPS threats/faults to model as well as analyze their inter-dependencies, inspired by attack-fault trees (AFTs). This investigation will include novel threat models in terms of AFTs and scalable techniques for formal analysis of AFTs; (ii) to collect threat intelligence on cyber-attacks in social VRLE deployments to detect critical anomaly events in real-time before the user safety gets compromised. This will lead to novel anomaly detection techniques based on machine learning models and statistical analysis techniques for the detection of cyber-attacks (single and multi-attack scenarios). The research outcomes validation features a real-world social VRLE that is hosted on available public cloud platforms.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.