Inter-system Biases Solution Strategies in Multi-Gnss Kinematic Precise Point Positioning

Mowen Li,Adria Rovira-Garcia,Wenfeng Nie,Tianhe Xu,Guochang Xu
DOI: https://doi.org/10.1007/s10291-023-01443-3
2023-01-01
GPS Solutions
Abstract:Estimating inter-system biases (ISBs) is important in multi-constellation Global Navigation Satellite System (GNSS) processing. The present study aims to evaluate and screen out an optimal estimation strategy of ISB for multi-GNSS kinematic precise point positioning (PPP). The candidate strategies considered for ISB estimation are white noise process (ISB-WN), random walk process (ISB-RW), constant (ISB-CT) and eliminated by between-satellite single-differenced observations (ISB-SD). We first present the mathematical model of ISB derived from the observation combination among different GNSSs, and we demonstrate the equivalence between ISB-WN and ISB-SD in the Kalman filter. In order to evaluate the performance of these four ISB solution strategies, we implement kinematic PPP with 1-month static data from 112 International GNSS service stations and two-hour dynamic vehicular data collected in an urban case. For comparison, precise orbit and clock products from the Center for Orbit Determination in Europe (CODE), GeoForschungsZentrum in Germany (GFZ) and Wuhan University (WHU) are employed in our experiments. The results of static tests show that the positioning accuracy is comparable among the four strategies, but ISB-CT performs slightly better in convergence time. In the kinematic test, there are more cycle slips than static test, and the ISB-CT improves the positioning accuracy by 15.7%, 38.9% and 63.2% in east, north and up components, and reduces the convergence time by 60.1% comparing with the other strategies. Moreover, both the static and kinematic tests prove the consistence among CODE, GFZ and WHU precise products and the equivalence between ISB-WN and ISB-SD strategies. Finally, more, i.e., the same amount of cycle slips as for the dynamic data, are artificially added to the static data to conduct the pseudo-kinematic test. The result shows that ISB-CT improves the positioning accuracy and convergence time by 19.2% and 24.4%, respectively.
What problem does this paper attempt to address?