Visibility Analysis of 3D Urban Space Navigation Satellites Based on Virtual Panoramic Obstruction Images

The visibility of satellites is an important parameter for evaluating and predicting the accuracy and credibility of global navigation satellite system positioning. Since the satellite signal is easily blocked, reflected, and scattered by buildings, overpasses, and trees, the visible satellite number dramatically changes in time and space. The three-dimensional (3D) urban model, point clouds, and panoramic images are used to calculate the visibility of satellites. However, the refined urban model requires abundant manual work and is expensive, while point clouds with the characteristic of large data volume have the problem of computing efficiency low. Also, the public panoramic images only cover road areas, which can not calculate the visibility of satellites in 3D urban space. The existing methods are impossible to achieve refined, lightweight, and fast computation of the visible satellite number. Therefore, we propose the satellite visibility calculation method based on constructing virtual panoramic obstruction images through digital surface model (DSM). The required DSM range of sampling points is selected using an adaptive method considering the GNSS satellite elevation mask angle to reduce data volume. Then, the virtual panoramic obstruction imagery is constructed based on the determined range DSM, which saves time for the real-time calculation of satellite visibility. Hence, the calculation efficiency is greatly improved. To verify the effectiveness and reliability of the proposed method, we collected the experiment field point clouds and constructed high-precision DSM for calculating satellite visibility of arbitrary locations. The experiments demonstrated that the proposed method provides an easy-to-use and high-precision solution to map the spatio-temporal visibility of satellites.