Robust combination of IGS analysis center GLONASS clocks

Abstract

The International GNSS Service (IGS) Analysis Centers (ACs) generate precise GNSS products by integrating tracking data from globally distributed IGS stations. The ACs’ products are further processed and combined for the positioning, navigation and timing users. We propose a robust least squares estimation for combining GLONASS clock products. Besides the difference in clock reference, systematic errors exist in the clock differences between different ACs, which show a linear trend and are completely removed. The clock combinations utilizing the final, rapid and ultra-rapid products of the IGS ACs were implemented in this study. The results of clock validation show that the agreement and stability of the newly generated combination products are better than those of most ACs. Furthermore, the impact of the different clock combination products on precise point positioning (PPP) is analyzed. Compared to the results of PPP using the products generated by the traditional combination strategy, the repeatability of GLONASS static PPP using the new clock combinations is 2.31, 2.95, and 5.62 mm, an improvement by 26.7, 28.9 and 20.6% in N, E and U respectively. The average RMS of GLONASS kinematic PPP using the new clock combinations is 1.20, 1.47 and 3.01 cm, an improvement by 67.5, 71.9 and 70.3% in N, E and U respectively. The improvement of the proposed strategy on PPP results is significant.