Combined BDS-2/BDS-3 real-time satellite clock estimation with the overlapping B1I/B3I signals

Abstract

With the deployment of the global BeiDou navigation satellite system (BDS-3) constellation completed, the positioning service performance is significantly improved relative to the regional system (BDS-2). As one of the key prerequisite products for precise point positioning (PPP), the combined BDS-2/BDS-3 real-time satellite clock estimation method is investigated in this study. Since the code hardware delays of BDS-2/BDS-3 satellites might be inconsistent for some receiver types, the inter-system bias (ISB) is considered in the combination although the overlapping B1I/B3I signal is employed. The combination processing is comprehensively evaluated in terms of satellite tracking coverage, clock precision, and convergence time. The experiment results indicate that the standard deviation (STD) precision of the estimated clocks for BDS-2 medium Earth orbit (MEO) satellites is improved by 52.0% from 0.150 ns to 0.072 ns due to the improvement of satellite tracking coverage. Introducing the ISB parameter as piece-wise constants improves the STD of all clocks from 0.076 ns to 0.067 ns and BDS-3 clocks achieve a significant improvement from 0.065 ns to 0.053 ns. The real-time clock estimation is demonstrated to be robust for the clock datum transformation, and it converges after one hour to a stable precision but needs a long time to achieve the best performance. The BDS real-time clocks are estimated with the proposed strategy to support real-time PPP based on the predicted satellite orbits. Results show that the clock STD is about 0.1 ns for most BDS satellites and the positioning accuracy of about 10 cm is achieved after the convergence time of about half an hour. The clock precision for BDS-2 geostationary Earth orbit (GEO) satellites is degraded to 0.303 ns due to the impact of poor orbit prediction errors. The exclusion of GEO satellites improves the positioning accuracy by about 15% and reduces the convergence time by 22–34%.