Cooperative Vehicle Awareness using C-V2X boards

In recent years, technological growth has given a significant boost to the development of the automotive sector. Inter-vehicular communications, as they occupy a prominent position among the objectives of car manufacturers, are the subjects of discussion in this thesis. The main standardization bodies that have worked on the implementation of technologies suitable for vehicular communication are the Institute of Electrical and Electronics Engineers (IEEE) and the European Telecommunication Standard Institute (ETSI). The first is known for the realization of the WAVE stack (Wireless Access in Vehicular Environment), while the latter for the ITS-G5 stack. The first aim of this thesis is to introduce the reader to the world of vehicular networks, focusing in particular on the two stacks mentioned earlier. The characteristics and peculiarities of WAVE and ITS-G5 are described, delving also into the details of the Cellular Vehicle-to-everything (C-V2X) technologies which are the main topic of this work. C-V2X, compared to other access layers (such as 802.11p), is an emerging technology that puts its basis on existing cellular networks and all its most important features are studied here. The main part of this work is related to the implementation of a cooperative awareness algorithm, running on top of an ETSI ITS-G5 stack over C-V2X, in the context of the 5G-CARMEN (5G for Connected and Automated Road Mobility in the European Union) European project. It deals with the creation of a 5G-enabled corridor providing a multi-tenant platform capable of supporting the automotive sector, making it more safe, green and intelligent. The part of the project on which we focused is related to a Cooperative Lane Merge (CLM) use case which gives the vehicles the ability to autonomously move between the lanes avoiding collisions, thanks to the coordination supported by V2X and 5G technologies. The work was performed with the aim of integrating the developed algoritm inside Qualcomm experimental C-V2X boards, on which a commercial, automotive-grade, ETSI ITS-G5 stack was loaded. The developed algorithm, called CAM GeoFilter (or Vehicle Local Dynamic Map), is the core of our work, it is implemented in a C development environment and deals with the classification of surrounding vehicles through the exchange of Cooperative Awareness Messages (CAMs). The implementation and analysis were carried out on two Linux virtual machines, containing all the development tools and enabling the emulation of vehicles as if they were equipped with the C-V2X boards. This work describes in detail the logic of the algorithm and the obtained results. Virtual machines allowed us to work on the stack and emulate it without the necessity of dealing directly with the boards and real vehicles, as one virtual machine represented the vehicle with the classification algorithm and the other any remote vehicle to be classified. A commercial Vehicle Simulator, which is in reality a full-fledged emulator, was the software of choise for the validation of the algorithm, as it allowed us to perform simulations and verify the reliability of the developed application.