Fast Prediction Method for GNSS Satellite Visibility Determination in Urban Canyons

Urban canyons always negatively affect the performance of global navigation satellite system (GNSS). Using 3D city models to predict GNSS satellite visibility is one of the most common methods concerning urban canyons. However, due to the extremely huge data of 3D city models, calculations involved in the prediction of GNSS visibility can be very time-consuming. To solve this problem, a method was proposed that was capable of quickly predicting the existence of urban buildings blocking GNSS satellite signals. This method reduced the burden of calculation and shortened the time for calculating by analyzing the direction of normal vector of the triangular patch, the relative positions of the triangular patch and the user, and the occlusion relation between two adjacent moments. A simulation of this method was conducted. The results show that the intersection calculation counts is only 7.21% of the traditional method, and the computing time 22.38%, while the average relative error is just 1.89%. In conclusion, this new method has a higher computing efficiency and higher accuracy.