Open Collections

Abstract

Connected Vehicle (CV) networks facilitate wireless data exchange between vehicles, infrastructure, and other devices (V2X). A reliable and clear wireless medium is required for the implementation of a CV network especially for safety-related applications. CV networks based on DSRC, ETSI ITS-G5 and C-V2X technologies are susceptible to interference and congestion from both non-CV devices that may be authorized to share the same band and unintentional emitters. Currently, there is no mechanism to detect such interference and congestion which may lead to unreliable communication. Thus, in this research we introduce the need, identify the options, provide a practical scalable architecture for large-scale interference and congestion monitoring system in CV networks, and deploy and validate a proof-of-concept (PoC) demonstration. The architecture comprises three layers: detection, reporting, and central authority. For the detection layer, we developed three IEEE-based models using the IEEE 802.11p Physical Layer Convergence Procedure (PLCP) and Physical Medium Dependent (PMD) state machines, and one C-V2X model based on Sensing-Based Semi-Persistent Scheduling (SB-SPS). For the reporting layer, we adapt the IEEE 1609 Security Credential Management System (SCMS), originally developed for DSRC, to enable secure reporting of spectrum misbehaviour. Devices (RSUs or OBUs) detecting an event can generate a Spectrum Misbehaviour Report (MBR). If sufficient evidence is collected, the report is forwarded through SCMS to a Spectrum Misbehaviour Authority. For the central authority, we designed and validated a reference Spectrum Misbehaviour Authority (SMA) capable of interference source localization. Our approach combines state-of-the-art localization algorithms with unsupervised learning techniques, eliminating the need for prior knowledge of interference sources. In summary, we have proposed detection of spectrum misbehaviour at the device level, reporting of spectrum misbehaviour via an already implemented system with minor modifications, and analyzing of the spectrum misbehaviour reports in a central unit. We have validated the feasibility of the detection mechanism in the lab environment as well as in real-world environment. We have defined a detailed architecture of the SMA and its components including database, data validation and authentication unit, data processing unit structure, and others. We finalized the design and implemented the operational PoC at the AURORA connected vehicle testbed.