Modern vehicles contain scores of Electrical Control Units (ECUs) that broadcast messages over a Controller Area Network (CAN). Vehicle manufacturers rely on security through obscurity by concealing their unique mapping of CAN messages to vehicle functions which differs for each make, model, year, and even trim. This poses a major obstacle for after-market modifications notably performance tuning and in-vehicle network security measures. We present ACTT: Automotive CAN Tokenization and Translation, a novel, vehicle-agnostic, algorithm that leverages available diagnostic information to parse CAN data into meaningful messages, simultaneously cutting binary messages into tokens, and learning the translation to map these contiguous bits to the value of the vehicle function communicated.
@article{arxiv.1811.07897,
title = {ACTT: Automotive CAN Tokenization and Translation},
author = {Miki E. Verma and Robert A. Bridges and Samuel C. Hollifield},
journal= {arXiv preprint arXiv:1811.07897},
year = {2018}
}
Comments
5th Annual Conference on Computational Science & Computational Intelligence (CSCI'18)