ECU-Secure: Characteristic Functions for In-Vehicle Intrusion Detection

AutorChevalier, Yannick; Rieke, Roland; Fenzl, Florian; Chechulin, Andrey; Kotenko, Igor
Datum2020
ArtConference Paper
AbstraktGrowing connectivity of vehicles induces increasing attack surfaces and thus the demand for a sophisticated security strategy. One part of such a strategy is to accurately detect intrusive behavior in an in-vehicle network. Therefore, we built a log analyzer in C that focused on payload bytes having either a small set of different values or a small set of possible changes. While being an order of magnitude faster, the accuracy of the results obtained is at least comparable with results obtained using standard machine learning techniques. Thus, this approach is an interesting option for implementation within in-vehicle embedded systems. Another important aspect is that the explainability of the results is better compared to deep learning systems.
KonferenzInternational Symposium on Intelligent Distributed Computing (IDC) <13, 2019, St. Petersburg>
ProjektBundesministerium für Bildung und Forschung BMBF (Deutschland)/16KIS0835
ReferenzKotenko, Igor et al.: Intelligent Distributed Computing XIII. Cham: Springer, 2020. (Studies in computational intelligence 868), pp. 495-504