Distributed Estimation Based Public Key Infrastructure for In-Vehicle Network Security

Automotive Electronic Control Unit (ECU) security is a challenging problem due to over-the-air updates, cloud services, peripherals from third parties, and other third-party interfaces. In these scenarios, alongside the security triad (confi-dentiality, integrity, and availability), entity identity also becomes crucial to secure vehicles. This paper presents distributed estimation for the public key infrastructure (DE4PKI) technique which enhances the PKI framework with an encryption scheme that uses distributed estimation and cipher text. The main idea is to establish identity using PKI whereas data encryption uses a symmetric key alongside distributed estimation of CPS ciphertext to verify attacked nodes. The technique provides multiple layers of security through dynamic graphs, CPS variables, and PKI framework. It helps identify attacked nodes and eliminate them without hardware interventions and re-issuance of certificates once the ECUs are recovered from the attack. The distributed estimation framework provides scalability and localizes the at-tacks. The proposed technique is demonstrated on a typical ECU topology in a vehicle.