Current device authentication between many smart devices requires users to manually input credentials, making device provisioning difficult and cumbersome. To address this challenge, we develop novel methods utilizing various environmental contexts (i.e., powerline noise, vibration, and electromagnetic radiation) to securely authenticate devices with minimal user-involvement. Our works have broad range of applicability to today’s emerging connected ecosystems, ranging from small personal devices to large building management devices. 

Security related challenges continue to impede broader adoption of secure vehicular networks. Current entity authentication mechanisms are not only computationally intensive, but also suffer from scalability issues that hinders timely authentication of increased volume of entities. This research aims to improve the security of vehicle-to-everything (V2P, V2V, and V2I) communication by developing light-weight sensor-based authentication scheme designed to rapidly authenticate diverse mobile entities within heterogeneous vehicular environments.

In this research, we address inherent challenges that exist in the current zero-interaction device authentication domain such as improving randomness of extracted key and providing efficient ways to adjust the authentication range. The developed methods not only effectively protect against various cyber threats, but also enhance user experience, allowing seamless integration into daily activities. Leveraging statistical analysis, we refine the process of generating keys from environmental data, ensuring high entropy and thus stronger security.