An improved method for detecting BeiDou signal-in-space anomalies from precise ephemerides

The signal-in-space (SIS) anomalies caused by satellites and control segments can greatly affect the reliability and safety of navigation and positioning users. The prior information associated with the failure of the Advanced Receiver Autonomous Integrity Monitoring (ARAIM) algorithm were obtained by the evaluation of SIS failure rates broadcasted with navigation ephemeris to investigate the integrity of navigation and positioning. For the existing ARAIM algorithm, the failure rate of satellites in the BeiDou Navigation Satellite System (BDS) is a conservative estimate, which is inconsistent with the actual SIS performance of BDS. Only the accurate detection of the SIS anomalies of BDS satellites can provide an effective reference to associate with failure. Therefore, to improve the accuracy of SIS anomaly detection for BDS satellites, and to provide higher integrity services for users, this study presents an improved method of SIS anomaly detection with precise ephemeris. The median method was used to detect a gross error in clock data before the calculation of clock datum, and the combination of an experience threshold and trimmed mean was used to determine the anomaly detection threshold. The feasibility and efficiency of the proposed method were also analyzed using data collected between 2015 and 2016. The detection results show that the constellation fault caused by erroneous clock data can be avoided, and the SIS anomalies can also be detected using the proposed method. Additionally, through the comprehensive tests performed in this study, it was found that from 2015 to 2016, the average accumulated duration of anomalies for BDS satellites was 10 h for geostationary orbit (GEO) and incline geosynchronous orbit (IGSO) and 55 h for medium earth orbit (MEO), respectively. These anomalies were primarily caused by the satellite clock.