Global navigation satellite system (GNSS) positioning in dense urban areas remains a challenge due to the signal reflection by buildings, namely multipath and non-line-of-sight (NLOS) reception. These effects degrade the performance of low-cost GNSS receivers such as in those smartphones. An effective three-dimensional (3D) mapping aided GNSS positioning method is proposed to correct the NLOS error. Instead of applying ray-tracing simulation, the signal reflection points are detected based on a skyplot with the surrounding building boundaries. The measurements of the direct and reflected signals can thus be simulated and further used to determine the user's position based on the measurement likelihood between real measurements. Verified with real experiments, the proposed algorithm is able to reduce the computational load greatly while maintaining a positioning accuracy within 10 metres of error in dense urban environments, compared with the conventional method of ray-tracing based NLOS corrected positioning.

中文翻译：

---

一种计算有效的基于距离的 3D 映射辅助 GNSS 与 NLOS 校正方法

由于建筑物的信号反射，即多路径和非视距 (NLOS) 接收，在密集的城市地区进行全球导航卫星系统 (GNSS) 定位仍然是一个挑战。这些影响会降低低成本 GNSS 接收器的性能，例如那些智能手机。提出了一种有效的三维（3D）映射辅助GNSS定位方法来纠正NLOS误差。不是应用光线追踪模拟，而是基于具有周围建筑物边界的天空图来检测信号反射点。因此可以模拟直接和反射信号的测量值，并进一步用于根据实际测量值之间的测量可能性来确定用户的位置。经真实实验验证，