Toward Achieving Fine-Grained Access Control of Data in Connected and Autonomous Vehicles
Jie Cui, Xuelian Chen, Jing Zhang, Qingyang Zhang, Hong Zhong
Abstract
A connected and autonomous vehicle (CAV) is often fitted with a large number of onboard sensors and applications to support autonomous driving functions. Based on the current research, little work on applications' access to in-vehicle data has been done. Furthermore, most existing autonomous driving operating systems lack authentication and encryption units. As such, applications can excessively obtain confidential information, such as vehicle location and owner preferences and even upload it to the cloud, threatening the security of the vehicle and the privacy of the owner. In this study, we propose a fine-grained access control scheme to restrict applications' access to data in CAVs (FGAC-inCAVs). First, we present a system model composed of the following elements: a trusted third party (TTP), which is a fully trusted authority; perception components like sensors, which can capture the road information (pictures, videos, etc.); and multiple applications. Then, a fast attribute-based encryption (ABE) is presented, and security analysis also shows it is secure against selective and chosen-plaintext attacks. Furthermore, we propose a key update scheme based on the Chinese remainder theorem (CRT). Finally, the theoretical analysis and simulation experiments demonstrate its feasibility and efficiency.