Revisiting attacker model for smart vehicles

Abstract

Because of the potential impact on user's life in cooperative automated safety applications, the security of Vehicle-to-X communication (V2X) is mandatory. However, the current attacker model used in literature is often too network-oriented, and it is unclear what realistic attacks could be. In this paper, we use the V2X data lifecycle to derive the attack surfaces. From this, we lay the foundations of a revisited attacker model, which details realistic attacks and identify appropriate countermeasures. We demonstrate that while the security of data processing, data at rest, and data in-transit is well-advanced, the security of meta-data and data acquisition requires extra attention by the research community.