Security Analysis of the TSN Backbone Architecture and Anomaly Detection System Design Based on IEEE 802.1Qci

Abstract

With the development of intelligent and connected vehicles, onboard Ethernet will play an important role in the next generation of vehicle network architectures. It is well established that accurate timing and guaranteed data delivery are critical in the automotive environment. The time-sensitive network (TSN) protocol can precisely guarantee the time certainty of the key signals of automotive Ethernet. With the time-sensitive network based on automotive Ethernet being standardized by the TSN working group, the TSN has already entered the vision of the automotive network. However, the security mechanism of the TSN protocol is rarely discussed. First, the security of the TSN automotive Ethernet as a backbone E/E (electrical/electronic) architecture is analyzed in this paper through the Microsoft STRIDE threat model, and possible countermeasures for the security of automotive TSNs are listed, including the security protocol defined in the TSN, so that the TSN security protocol and the traditional protection technology can form a complete automotive Ethernet protection system. Then, the security mechanism per-stream filtering and policing (PSFP) defined in IEEE 802.1Qci is analyzed in detail, and an anomaly detection system based on PSFP is proposed in this paper. Finally, OMNeT++ is used to simulate a real TSN topology to evaluate the performance of the proposed anomaly detection system (ADS). As a result, the protection strategy based on 802.1Qci not only ensures the real-time performance of the TSN but can also isolate individuals with abnormal behavior and block DoS (denial of service) attacks, thus attaining the security protection of the TSN vehicle-based network.

IEEE 802.1 Zaman Hassas Araç-İçi Haberleşme Ağları için Siber Güvenlik: Güncel Gelişmeler ve Hizmet Reddi Saldırılarının Etki Analizi

“…Emphasizing that the original specification for CAN lacks a central monitoring mechanism, Häckel et al [45] offers an architecture integrating TSN and SDN, in which all real-time traffic transmitted in an in-vehicle network is centrally monitored without increasing the network delay in order to prevent malicious traffic. Meyer et al [46] performs anomaly detection by using TSN and SDN technologies together in an invehicle communication network, which relies on the filtering capabilities of Per-Stream Filtering and Policing (PSFP) specified [47] proposes an anomaly detection system for invehicle networks with TSN support relying on PSFP specified in 802.1Qci in order to block the messages exceeding a customized maximum service data unit size. The authors experimentally show that the proposed system successfully identifies four different abnormal traffic events in relation with DoS attacks.…”

Toward Deterministic Communications in 6G Networks: State of the Art, Open Challenges and the Way Forward

“…Several other works considered 802.1Qci's PSFP, for instance, [80] presented a survey that compares works related to AVB and TSN security. In [81], DoS attacks were identified as the biggest hidden danger that needs to be considered in the design of TSN architectures. A PSFP-based anomaly detection system was designed and evaluated by the authors.…”

Automotive Ethernet architecture and security: challenges and technologies

“…Yet, the protection system of the TSN protocol is seldom investigated. Luo et al [46] analyzed the protection of the TSN automotive Ethernet as a pillar of E/E design; the authors utilized the Microsoft STRIDE threat model, and defenses for the protection of automotive TSNs were reviewed, the protection method PSFP defined in IEEE 802.1Qci was analyzed thoroughly, then they proposed anomaly detection system based on PSFP. After all, OMNeT++ was utilized to imitate a true TSN topology to assess the functioning of the suggested ADS.…”