The following text describes in detail of the subsystems composing the second RAINBOW use case.
Connected Vehicle: Vehicle equipped with a C-V2X communication device and a V2X client.
Figure 1: Connected Vehicle
The Vehicle performs data acquisition from the on-board CAN Bus and dispatches it to a cloud node using ETSI CAM messages through the Uu (and PC5) interface and the AMQP protocol. The Vehicle also receives ETSI DENM messages from Fog Nodes via PC5 or Uu interface. The messages of interest for the vehicle are the Hazardous Location Notification (DENM) messages with different cause code depending on the type of hazard. When the vehicle receives DENM messages from the Fog Node, it parses and displays them on the HMI depending on their relevance.
The main components on the Vehicle are:
- PC5 and Uu protocol stacks: each manages the UU and PC5 communication layer.
- AMQP Client: it establishes a logical communication with the AMQP broker on the cloud node and it extracts and encapsulates V2X messages (CAM, DENM).
- V2X Management: it handles the transmission and reception of CAM and DENM messages.
- HMI: it displays the messages to the end user according to relevance policies.
From the privacy and security point of view, the use case will follow the PKI architectures defined by ETSI ITS and IEEE to secure all V2X communications. For privacy protection purposes, certificates have a reasonably limited validity period in order to limit the reusability in case of revocation. Given the frequent transmission of CAM messages from vehicles, including location information, this can create a scalability problem with a centralized PKI-based approach [1].
Fog Node: RSU node with PC5 interface, equipped with one or more IP cameras and a GPU AI computer.
Figure 2: Fog Node
The Fog platform includes a platform equipped with IP cameras, able to perform AI computer vision tasks and to communicate with vehicles both through PC5 communication and AMQP messages (via cloud). On the computer vision side, the platform manages the video stream data gathering and forwards them to the MEC node. Moreover, it can run AI detection algorithms based on Rainbow orchestrator decisions. On the communication side, the Fog platform is in charge of broadcasting hazardous event notification through ETSI DENM messages (via PC5).
MEC Node: A node that simulates an Edge Server typically installed inside a mobile or road operator network.
Figure 3: MEC Node
The MEC platform is typically hosted in a mobile or road operator network. The MEC node can run the AI detection algorithms thanks to the video streams received by the RSU. The MEC node can notify hazardous situations to the RSU (that will broadcast them directly to vehicles via PC5) and to the cloud node (where they can be retrieved via Uu by vehicles interested in notifications from a certain area).
Citizen App: App that allows Citizen to alert the municipality about Hazardous situations on the road.
Figure 4: Citizen App
The Citizen App publishes Hazardous situations on the road on the City Cloud Node.
City Cloud Node: Aggregates contextual traffic data coming from vehicles, local traffic control centre or citizens.
Figure 5: Cloud Node
References
[1] I. K. Thanassis Giannetsos, “Securing V2X Communications for the Future: Can PKI Systems offer the answer? In Proceedings of the 14th International Conference on Availability, Reliability and Security (ARES ’19). Association for Compu,” in 14thInternational Conference on Availability, Reliability and Security (ARES ’19).Association for Computing Machinery, New York, NY, USA, 2019.