Cyber Security of the Railway wireless system: detection, decision and Human-in-the-Loop

Christophe Gransart (1, 2) , Virginie Deniau (1, 2) , Eric Pierre Simon (3, 2, 4) , Anthony Fleury (5, 6, 7, 2) , Stéphane Lecoeuche (5, 6, 7, 2) , Emilie Masson (2) , Patrick Millot (8, 2)

1 IFSTTAR/COSYS/LEOST - Laboratoire Électronique Ondes et Signaux pour les Transports
2 IRT Railenium - L'Institut de Recherche Technologique (IRT) de la filière Ferroviaire
3 TELICE - IEMN - Télécommunication, Interférences et Compatibilité Electromagnétique - IEMN
4 IEMN - Institut d’Électronique, de Microélectronique et de Nanotechnologie - UMR 8520
5 CERI SN - IMT Nord Europe - Centre for Digital Systems
6 IMT Lille Douai - Ecole nationale supérieure Mines-Télécom Lille Douai
7 Université de Lille
8 LAMIH - Laboratoire d'Automatique, de Mécanique et d'Informatique industrielles et Humaines - UMR 8201

The networks used in the Railway domain are usually heterogeneous, not enough protected and not fitted to the usual Cyber Security requirements in terms of sustainability, protection and attack detection. Furthermore, the quick evolution of the telecommunication means, the threats and the sustainability aspects have to be taken into account in order to protect the Railway system. The paper presents the first contributions on Cyber Security for railways that can be divided into three main aspects dealing with the Cyber Security of the wireless part of the railway communication system: detection, decision and Human-in-the-Loop. Part of the work will be devoted to the development of an Open Pluggable Framework (OPF). The OPF is a software framework based on automation principles. It monitors the environment, then some algorithms detect abnormal behaviours, and next, OPF decides which reaction to take and finally apply this action (e.g. an alarm or a reconfiguration). The last part " human in the loop " aims at answering the questions: what happens if the automatic countermeasures fail and how the driver can cope with the attack consequences. It consists in placing professional drivers and OCC operators (Operational Control Center) in a realistic simulator and playing scenarios involving attacks. Scenarios are deduced of a vulnerability analysis and human-system resilience is studied experimentally in observing the human abilities to detect, to understand and to react to (simulated) cyber-attacks, or to their consequences. A preliminary methodology is proposed and discussed through a concrete case study.

Mots clés: cyber-attacks detection; Software Defined Radio (SDR); classification algorithms; Train-tram Simulator; scenarios of threats; analysis of driver behaviour

Domaines: Traitement du signal et de l'image [eess.SP]

Christophe Gransart, Virginie Deniau, Eric Pierre Simon, Anthony Fleury, Stéphane Lecoeuche, et al.. Cyber Security of the Railway wireless system: detection, decision and Human-in-the-Loop. TRA 2018 (Transport Research Arena), Apr 2018, Vienna, Austria. 10p. ⟨hal-01854846⟩