Defensing the malicious attacks of vehicular network in runtime verification perspective

Authors:
  • XZXiang Zhang
    Xiang Zhang
  • WFWufeng Feng
    Wufeng Feng
  • JWJiangtao Wang
    Jiangtao Wang
  • ZWZhenghui Wang
    Zhenghui Wang
Publisher website
Google Scholar
Abstract
The Controller Area Network (CAN) protocol, which is frequently employed in today's automobiles, offers the communication standards between different Electronic Control Units (ECUs) on vehicular system[1]. However, this protocol does not consider some of the security issues like authentication and access control, which make CAN vulnerable to kinds of attacks[2]. This paper mainly focuses on some of the most common causes of attacks on vehicular network, and propose a Runtime Verification[3] (RV) based framework for defensing malicious attacks against CAN protocols. Especially the attacks launched form exposed interfaces of vehicular systems, such as Bluetooth, Telematics and On Board Diagnostic (OBD) port. Briefly, we adopt timed automaton to describe ECU behaviors and create security specifications to detect illegal behaviors on CAN bus. Then we develop a framework based on Copilot[4] to implement the detection at runtime. In addition, our implementation is also able to defense the malicious attack before it impacts the vehicular system.

This publication has 11 references indexed in Scilit: