Railway GNSS Multipath Error Modelling Approach with both Train-Side and Operational Environment Characterization

Railway transportation systems have high accuracy and high integrity demands for safe localization. In the future, railway signaling is expected to rely on onboard sensors like Global Navigation Satellite Systems (GNSS) in order to reduce installation and maintenance costs. GNSS position determination can be however highly degraded because of the presence of multipath caused by the train roof and the challenging railway environment. In this work, we propose to model multipath error in the pseudorange as a combination of the multipath introduced, on one hand, by the vehicle structure and antenna installation and, on the other hand, by the additional reflections of surrounding buildings and objects. First, we derive a safe multipath error model for the antenna installation on the train roof. Second, we use the obtained error model as a reference model to normalize and characterize multipath in along the railway tracks. The role of different Fault Detection and Exclusion (FDE) algorithms in ensuring model compliance is also evaluated. Finally, the methodology is evaluated and tested with real railway dataset. The characterization of both the permanent train multipath and the operational environment multipath contributes to a safe and robust complete GNSS error description for safety critical navigation in the railway environment.