Search

Cyber Physical Systems are characterized by a strong interaction among networking, sensing, and control functionalities. Moreover, the recent advent of Internet of Things extended their information sharing capability. However, the interaction between Internet and Cyber Physical Systems requires increased efforts for guaranteeing the security of connected systems. In the industrial field, the problem becomes more complex due to the need of protecting a large attack surface while guaranteeing system availability and real-time response to the detection of threats. In this contribution, we deal with the injection of tampered data into the communication channel with the aim of modifying the status of the physical system. To cope with this attack, we design a secure control system able to detect the injection of tampered data by coding the output of the measurement systems. The proposed approach is based on the use of permutation matrices, whose scheme varies upon a secret pattern obtained exploiting the Fibonacci $p$ -sequences. The detection strategy is compliant with the time delay constraints typical of a Cyber Physical System. An analysis of the security performances of the proposed system is presented along with the experimental proof of its effectiveness.

Nowadays, Global Navigation Satellite Systems are the main source for high accuracy positioning and timing. For this reason, they are essential both for everyday activities and services, and for the industrial and critical infrastructure sectors. Moreover, the spread of increasingly autonomous vehicles results in strict accuracy and integrity requirements. This leads to the need for additional infrastructure to send corrections to the end users and mitigate the measurement errors, the Augmentation Networks. However, due to the increasing exploitation of localization functionalities, the Augmentation Networks could become a primary target for attackers resulting in a high financial and safety cost. Among the possible attacks, spoofing, that is the generation of a fake satellite signal which is seen as genuine by the receiver, is one of the most powerful and tricky. In this contribution, a detection and mitigation strategy for Augmentation Network spoofing attacks is proposed. We introduce two attack models and present a technique based on K-means clustering to counteract them. More in details, our approach is based on the computation of the number of clusters formed by the Augmentation Network corrections. Starting from the hypothesis that under nominal conditions only one cluster is present, the effects of the attacks on the clustering procedure are analyzed, and several attack simulations are performed to evaluate the algorithm performances. The proposed method has been compared both to an Augmentation Network attack detection technique, and to a receiver-level spoofing mitigation approach, showing comparable or better performances. Moreover, to the best of our knowledge, this is the first work addressing mitigation for spoofing attacks which target an Augmentation Network.