PRELIMINARY ANALYSES OF BEIDOU SIGNAL-IN-SPACE ANOMALY SINCE 2013

As BeiDou navigation system has been operational since December 2012. There is an increasing desire to use multiple constellation to improve positioning performance. The signal-in-space (SIS) anomaly caused by the ground control and the space vehicle is one of the major threats to affect the integrity. For a young Global Navigation Satellite System, knowledge about SIS anomalies in history is very important for not only assessing the SIS integrity performance of a constellation but also providing the assumption for ARAIM (Advanced Receiver Autonomous Integrity Monitoring). In this paper, the broadcast ephemerides and the precise ones are pre-processed for avoiding the false anomaly identification. The SIS errors over the period of Mar. 2013-Feb. 2016 are computed by comparing the broadcast ephemerides with the precise ones. The time offsets between GPST (GPS time) and BDT (BeiDou time) are estimated and removed by an improved estimation algorithm. SIS worst-UREs are computed and a RMS criteria are investigated to identify the SIS anomalies. The results show that the probability of BeiDou SIS anomalies is in 10 level in last three years. Even though BeiDou SIS integrity performance currently cannot match the GPS integrity performances, the result indicates that BeiDou has a tendency to improve its integrity performance. * Corresponding author 1. INTODUCTION BeiDou Navigation Satellite System (BDS) and timing in the service in the Asian-Pacific region since the late 2012. With the modernization of GPS, the fully-operation constellation of GLONASS, and the developing of Galileo system, there is an increasing desire to use the multi-constellation Global Navigation Satellite System (GNSS) to enhance position accuracy, continuity, integrity and availability. With access to a large number of the multi-frequencies pseudo-ranges, it is expected that the GNSS will support navigation for very highintegrity-demanding users, such as ARAIM(Advanced Receiver Autonomous Integrity Monitoring )(Blanch et al., 2015, Blanch et al., 2016) users in the precision approaching phases of aviation flights. For real-time GNSS positioning users, the broadcast navigation message including ephemeris data, are used to calculate the satellite orbits and clock corrections which indicate the time of the constellation. Therefore, differences between the orbits /clocks derived from the broadcast ephemerides and the truth, is called satellite signal-in-space Error (SISE). SISE is the mostly concerned element which determines the positioning accuracy and integrity. For stand-alone GNSS users, if a SISE is larger than a threshold (rigid definition given later), this SISE is call an anomaly. If the SIS anomaly happened without warning, it could result in a hazardous misleading. The SIS anomalies are caused by either the ground control or the space vehicle (Walter and Blanch, 2015). A typically example (GPS WORLD, 2014) of GNSS SIS anomaly is the one that GLNOASS simultaneous multiple SIS anomalies occurred on April 2 2014, lasting for 11 hours, from just past midnight until noon Russian time (UTC+4). The reason for system completely failure is that bad ephemerides were uploaded to satellites. It resulted in position solution was off by more than 50km. Many studies have analysed the broadcast ephemeris accuracies of BeiDou (Montenbruck et al., 2015, Hu, 2013, Chen et al., 2013), GPS (Montenbruck et al., 2015, Heng et al., 2011, Heng et al., 2012b) and GLONASS (Heng et al., 2012c). All the above studies are focused on the nominal SISE (satellite anomaly free) behaviours in a statistical sense. In practice, unfortunately, the signal-in-space anomalies occasionally occur, which result in the User Range Errors (URE) in tens of meter or even up to tens of kilometres. (Heng et al., 2012b) shows that 3275 potential SIS anomalies of GPS are found from 2000 to 2010. The total hour of potential SIS anomalies per year has been improving over the period. The experimental probability of GPS anomalies is supposed -5 10 /sat/hour (GPS-Galileo Working Group C, 2015). For GPS, there are no simultaneous multiple anomalies have occurred since 2004. Heng (2012a) identifies 192 potential SIS anomalies of GLONASS over the period of Jan 1, 2009 to Aug 11, 2012 and the anomaly probability has improved from -3 10 level to -4 10 during the period. For stand-alone ARAIM users, a SIS anomaly could result in the hazardous misleading position solution. Therefore , the knowledge, in term of anomaly probability and multiple simultaneous anomalies, is very important for the high integrity demanding users. Furthermore, the knowledge is expected to broadcast to users in integrity Support Message (ISM) (GPSThe International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLI-B1, 2016 XXIII ISPRS Congress, 12–19 July 2016, Prague, Czech Republic This contribution has been peer-reviewed. doi:10.5194/isprsarchives-XLI-B1-517-2016 517 Galileo Working Group C, 2015, Walter et al., 2014) for providing the a priori stochastic model for ARAIM. BeiDou System has been officially operational for more than three years. However, BDS anomaly has not been analysed yet. In this paper, the BDS broadcast ephemerides are downloaded and pre-processed to filter out the duplications and screen out the errors caused by receiver data conversion software, and etc. SISEs are calculated by comparing the broadcast ephemerides and the precise ones (produced by Wuhan University) (Zhao et al., 2013). Then, algorithm for estimating and removing the time-offset between GPST and BDT are investigated. RMS (Root Mean Squared error) based criteria is developed to identify the SIS anomalies. Finally, the BDS anomalies over the three year from Mar 1, 2013 to Feb 29, 2016 are presented and discussed. 2. DATA SOURCE AND PROCESSING