Methodology for Selection of Optimal Satellite Constellation for Precise GNSS Positioning

Abstract

Continuous advancements in GNSS systems, apart from the broadly used GPS, have led to the development of other satellite systems (Galileo, BeiDou, GLONASS), which have significantly increased the number of available satellites for GNSS positioning applications. However, GNSS satellites’ redundancy and potential poor GNSS satellite signals can negatively affect the GNSS’s positioning speed and accuracy. On the other hand, selecting high-quality GNSS satellite signals by retaining a sufficient number of GNSS satellites can enhance the GNSS’s positioning performance. Various methods have been applied for satellite selection. However, the selected satellites via these methods could be insufficient for precise positioning. These methods were designed to select satellites for Single Point Positioning (SPP) solution with meters level accuracy. This is because these satellite selection methods only consider SPP factors. To choose satellites for precise positioning, new satellite selection method was developed considering Precise Point Positioning (PPP) factors. Various satellite selection methods were investigated and assessed to developed new satellite selection method that fit PPP- satellite selection. To test the performance of the new satellite selection method, several experiments have been conducted. Satellites were selected for various PPP applications: i) static open-sky, ii) kinematic open-sky, and iii) static at multipath conditions. In addition, satellites were selected in all possible subset sizes, i.e., they were selected in subsets with all possible numbers of satellites. According to experiment results, the efficiency of the new satellite selection method was excellent at low multipath environment regardless of PPP modes: static or kinematic. The positioning accuracy of original PPP, using all satellites, was achieved by multiple sizes of selected satellite subsets by the selection method. It was also obtained by selected satellites with different subset sizes. In terms of high multipath environments, the performance of the satellite selection method was less effective. The positioning accuracy of the original PPP was not achieved by the selected satellite subsets although they provided positioning accuracy close to the original PPP with a few satellites. This achievement is quite promising although the new satellite selection method needs further improvement.