Pietro Morri - Communication systems
Date: - Location: Eurecom
Abstract: The ongoing evolution of connected and automated mobility increasingly calls for localization capabilities that surpass those of conventional GNSS, especially in urban scenarios where satellite visibility is obstructed. This seminar presents an experimental testbed developed within the framework of the NextGenerationEU initiative, as part of the MOST National Center for Sustainable Mobility — Spoke 9 "Urban Mobility". We investigate the potential of 5G physical reference signals—specifically Sounding Reference Signals (SRS) and Positioning Reference Signals (PRS)—to enable high-accuracy vehicular localization in smart road environments. Two advanced driver assistance system (ADAS) use cases are highlighted: vulnerable road user (VRU) protection and automated parking. The testbed integrates automated vehicles, road-side sensing units, Software Defined Radios (SDRs), commercial-grade ITS-G5 and 5G modems, and a fully open-source 5G stack (RAN and core). This configuration supports flexible experimentation across both network-based and UE-based positioning paradigms. The seminar begins with a technical overview of the testbed architecture, detailing the deployment of physical and logical infrastructure components. We describe how positioning and communication layers are orchestrated to support the selected ADAS scenarios. We then focus on network-based localization, presenting both the theoretical framework and experimental implementation of uplink-based positioning using SRS. Specific attention is given to synchronization strategies across distributed base stations, limitations posed by geometric constraints (e.g., single-cell TDOA configurations), and hybrid techniques that fuse time-based and angle-based measurements, such as AoA/range fusion. In the UE-based localization section, we discuss onboard ego-positioning methods, comparing geometric approaches with signal-processing-based techniques. Topics include carrier-phase ranging and PRS-based radar-like sensing, with an emphasis on the implementation challenges and feasibility within open-source UE platforms. The final part introduces a prototype for cooperative localization and sensing, which fuses perception and positioning data through Implicit Cooperative Positioning (ICP) algorithms. By leveraging distributed situational awareness between vehicles and infrastructure, this multi-agent approach demonstrates significant potential in enhancing localization robustness and precision. Short bio: Pietro Morri received the bachelor’s degree in Electronics Engineering and the master degree in Telecommunication Engineering from Politecnico di Milano in 2018 and 2022 respectively. In 2023 he started a PhD in Information Technology at Politecnico di Milano focusing on 5G localization and sensing over V2X networks. His research is framed within the National Center for Sustainable Mobility (CN MOST) working towards the design and implementation of smart road infrastructures to develop and test Advanced Driver-Assistance Systems.
