An approach for defining faults in MIL-STD-1553 bus in an Avionics Architecture

Sagirkaya H., SCHMİDT Ş. E.

42nd IEEE/AIAA Digital Avionics Systems Conference, DASC 2023, Barcelona, İspanya, 1 - 05 Ekim 2023 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Doi Numarası: 10.1109/dasc58513.2023.10311103
  • Basıldığı Şehir: Barcelona
  • Basıldığı Ülke: İspanya
  • Anahtar Kelimeler: avionics, machine learning, MIL-STD-1553, simulation
  • Orta Doğu Teknik Üniversitesi Adresli: Evet

Özet

This paper presents a detailed and realistic simulation environment for the very widely used MIL-STD-1553 communication protocol. We then propose a machine-learning approach for detecting malicious traffic and preventing faults caused by cyber-attacks. Our approach utilizes clustering according to the features of a given dataset. We demonstrate our approach with the presented simulation environment in this paper. To this end, we construct a testbed to implement various scenarios such as spoofing and DoS attacks. We generate message traffic with unauthorized RT messaging. We apply our anomaly detection approach to our traffic and present an analysis of the results.