Multiepoch Grid-Based 3DMA Positioning in Dense Urban Canyons by Optimizing Reflection Modeling

Urban positioning with global navigation satellite systems (GNSSs) poses a significant challenge due to signal reflections, diffractions, and obstructions caused by surroundings. These factors introduce multipath interference and non-line of sight (NLOS) signal reception, leading to a degradation in positioning accuracies up to tens of meters. The growing accessibility of 3-D city models offers an opportunity for modeling and analyzing the effects of multipath and NLOS on GNSS signals. 3-D mapping aided (3DMA) methods have shown effectiveness in positioning improvement by correcting NLOS delays in preliminary research. However, without consideration of diffractions, which frequently occurs in urban canyons, the usable satellites from simulation are significantly reduced in their methods. Moreover, the conventional positioning scheme lacks robustness when receivers suffer from unmodeled errors and dynamic interference. In this contribution, we propose an enhanced 3DMA ray-tracing (RT) algorithm (RT-based grid weight smoothing and clustering, RT-GWSC). This method incorporates a reflection-diffraction (RD) model, and it introduces a positioning scheme with multiepoch weight smoothing and density-based spatial clustering of applications with noise (DBSCAN) clustering approach. The performance has been evaluated with different urban environments in both static and kinematic experiments. The results demonstrated that the RD model outperforms the conventional reflection model, with a significant improvement in sensitivity of NLOS prediction from 53.7% to 90.7% and overall accuracy from 86.3% to 96.8%. Overall, the RT-GWSC method could reduce outliers in positioning stage and achieve positioning accuracies better than 15 m in all tested environments, outperforming raw GNSS outputs with improvement rates (IRs) over 73%, while the conventional RT approach was found ineffective in certain complicated environments.