Initial Positioning Assessment of BDS New Satellites and New Signals

Abstract

With the official announcement of open service since the end of 2018, the BeiDou navigation satellite system (BDS) has started to provide global positioning, navigation and timing (PNT) services. Thus, it is worth assessing the positioning service of new BDS satellites and signals. In this paper, we comprehensively assess the system status and the global positioning performance of BDS regarding single point positioning (SPP) and real-time kinematic (RTK) performance. Results show that the signal in space range error (SISRE) of BDS-3 satellites is superior to that of BDS-2 satellites, showing an overall accuracy of 0.71 m versus 0.97 m, which is competitive with GPS and Galileo. With the contribution of BDS-3, the number of global average visible satellites has increased from 5.1 to 10.7, which provides a mean global position dilution of precision (PDOP) value better than 6 at 99.88% and the mean availability of basic PNT performance is also improved from 35.25% to 98.84%. One week of statistical results from 54 globally distributed international GNSS service (IGS) stations show that the root mean square (RMS) of SPP accuracy is 1.1 m in horizontal and 2.2 m in vertical, which is at the same level of GPS. The new B1c and B2a signals show a smaller observation noise than B1I, and SPP performance of B1c is similar to that of B1I. However, the positioning precision is slightly worse at the B2a frequency, which may be due to the inaccurate BDS ionosphere correction. As for short baseline RTK, baseline accuracy is also improved due to the increased number of new BDS satellites.

Calibration and Impact of BeiDou Satellite-Dependent Timing Group Delay Bias

“…For the SISRE performance of BDS-3, Yang et al [14] first reported an average value of 0.44 m using 4 days' worth of data from eight BDS-3 satellites by comparing the broadcast ephemeris and precise results from the International GNSS Monitoring and Assessment System (iGMAS). A comprehensive comparison with 2 months' worth of data from the 18 BDS-3 satellites showed that the SISRE of BDS-3 was 0.71 m after TGD correction and 0.57 m after DCB correction [15]. Therefore, except for the apparent TGD bias between BDS-2 and BDS-3, a dispersion bias of the TGD among BDS-3 satellites was also indicated.…”

“…As an MGEX analysis center, Wuhan University started to generate precise orbit and clock (WUM) of BDS-3 from 2019 (ftp://igs.ensg.ign.fr/pub/igs/products/mgex/). When comparing the precise products with the BDS broadcast ephemeris, the difference in the time and coordinate reference system, satellite antenna offset correction, and clock offset correction should be considered [15]. It should be pointed out that, different from the traditional B1I/B2I ionosphere-free (IF) combination, the WUM products apply the B1I/B3I IF combination for precise orbit determination, as the B2I signal is not supported by BDS-3 satellites [16].…”

“…We can clearly see that positioning improvement outside the East Asia region is more evident. One reason for this phenomenon is that with the decrease of visible satellites [15], the contribution of the TGD bias correction becomes more significant. The other reason may due to the more scattered TGD bias of BDS-3 satellites, as indicated in Figure 6.…”

BDS Satellite-Based Augmentation Service Correction Parameters and Performance Assessment

“…where ρ i brd,s is the geometric distance from station to satellite computed from the broadcast ephemeris, and dρ is the line of sight observation correction converted from the SBAS satellite orbit correction; δt i brd is the satellite clock calculated from the broadcast ephemeris, and ∆t i is the real-time SBAS satellite clock correction (in the form of ionospheric-free combination); STD i m,s is the slant tropospheric delay using the same model as in the BDS SBAS; ∆Φ i is the real-time SBAS partition comprehensive correction of carrier-phase, respectively. It is worth noting that the BDS broadcast clocks are referred to the B3 frequency, the timing group delay (TGD) parameters should be applied [23] in the dual-frequency ionospheric-free combination. Meanwhile, as the partition comprehensive correction contains most parts of tropospheric residuals within a certain area, it is recommended that there is no necessity to estimate the residual tropospheric delay considering the accuracy requirement of decimeter level [19,20].…”

Comparison of three widely used multi‐GNSS real‐time single‐frequency precise point positioning models using the International GNSS Service real‐time service

“…As for RT clocks, they could be predicted using polynomial fittings for a short time [2]. To remove temporal/spatial reference discrepancies between RT and precise products, key issues need to be taken into account: (i) The GBM products refer to the satellite centre of mass, whereas the CLK93 products adopt the antenna phase centre (APC) of the satellite, thus the satellite phase centre offset (PCO) should be corrected using igs14.atx [31] (ii) The clock products of CLK93 and GBM have the different clock datum, a medium GLK93‐minus‐GBM value of all satellite clocks at each epoch is calculated as a systematic bias to eliminate this difference [32].…”