CARDIAN: A Context Aware Cybersecurity System for Real Time Diagnostic Intrusion Detection Using a Probabilistic Approach with Bayesian Network

Smart Mobility is a concept for achieving sustainable city development. The term encompasses a series of elements: technology, mobility infrastructures (parking, charging networks, signage, vehicles), mobility solutions (including new mobility models) and people. To date, this paradigm is strongly linked to that of the IoT and Smart Vehicles as these describes the crucial role of the emerging pervasive technologies all around the world. In this context, the automotive sector plays a major role as vehicles have become active tools of the information network and are increasingly connected and automated. All this gives rise to new perspectives also of vulnerability as they are natively objects created not to be connected to the internet and therefore adapted to this way. One of the biggest challenges today is to obtain cyber-safe vehicles against cyber-attacks. This article describes a prototype solution (CARDIAN) which, through a real time diagnostic analysis algorithm, is able to understand if there is an attack in progress or in the internal communication network between the various Electronic Control Units (ECU) of the vehicle. Through a probabilistic approach, in fact, the CARDIAN is able to understand anomalous situations with a certain precision and to pilot any corrective actions. The first experimental results show how the model works correctly in a simulated environment.