Antoine CHEVROT : "Detection of Contextual Anomalies in Air Traffic Data Using Neural Network Models"

Abstract of PhD works :To face the ever-growing number of aircraft flying in the world airspace, the Air Traffic Control (ATC) needs to adapt and propose new technologies that are both cheaper to produce and more accurate, sometimes at the expense of the cybersecurity of its systems. The Automatic Dependent Surveillance-Broadcast (ADS-B) protocol is one of the latest compulsory advances in air surveillance and perfectly depicts this tendency: cheaper to maintain and to implement thanks to inexpensive transponders installed in each aircraft broadcasting their GPS information but way more vulnerable than older technologies. This is due to the change of paradigm that the ADS-B embodies. Older technologies like primary and secondary radars were watching the sky using powerful antennas tracking flights in a given area while with ADS-B, radars do not scan the air but rather directly receive information sent by each aircraft. For this reason, and because of the absence of encryption and authentication in the protocol, it is particularly vulnerable to False Data Injection Attacks (FDIA). FDIAs are messages either created, modified, or deleted by malicious individuals with the intent to disrupt traffic management. [...]

Jury Composition :

Pr. Ivan Martinovic (Reporter), Professor - University of Oxford
Pr. Pietro Michiardi (Reporter): Professor - Eurecom
Dr. Arnaud Gotlieb (Reviewer): Chief Research Scientist au SIMULA LAB
Pr. Olga Kouchnarenko (Reviewer): Professor - Université de Franche-Comté
Dr. Xavier Olive (Reviewer): Research Scientist à l'ONERA
Pr. Bruno Legeard (PhD co-director ): Professor - Université de Franche-Comté
Dr. Alexandre Vernotte (PhD Co-director): Postdoc - Université de Franche-Comté

Localization : Amphithéâtre J.J Gagnepain, Bâtiment TEMIS - FEMTO-ST, 15B, Avenue des Montboucons 25000 Besançon