Performance assessment of multiple ambiguity resolution methods in single LEO precision orbit determination

Low Earth Orbit (LEO) satellites play a vital role in fields of gravity field recovery, meteorological detection, and Precise Point Positioning (PPP) of augmented Global Navigation Satellite Systems (GNSS), which makes the precise orbit determination (POD) of LEO satellites the key to their mission-execution. Ambiguity resolution (AR) is an effective method to improve the orbit accuracy of LEO satellites, among which the Single Difference (SD) AR method requires GNSS hardware bias products. Besides, Track-to-Track (T2T) AR and Single Difference Track-to-Track (SDT2T) AR can be used as the alternative methods in the scenarios without any GNSS bias products. In this study, the reduced dynamic POD (RDPOD) of the GRACE-FO and SWARM constellations from day 001 to 150 of 2020 was performed using the SD AR, T2T AR and SDT2T AR methods. The results showed that the performance of T2T AR is inferior to that of SDT2T AR and SD AR due to the gradual variability in the phase fractional bias of satellites and receivers. By analyzing the relationship between the success rate of T2T AR and the time interval between adjacent ambiguities, it is found that the success rate of T2T AR is higher when the time interval is within 30 min. Therefore, it can be assumed that the fractional cycle biases of the receivers remains constant within 30 min because the performance of T2T AR highly depends on the stability of the fractional cycle biases of the receivers. With impacts caused by the receiver fractional biases eliminated, SDT2T AR significantly outperforms T2T AR and achieves comparable performance with SD AR. Therefore, in the cases that GNSS bias products are unavailable, SDT2T AR is recommended as the preferred option for single LEO POD.