An improved weighting strategy for tropospheric delay estimation with real-time single-frequency precise positioning

In recent years, single-frequency precise point positioning (SF-PPP) has taken increasing attention from the GNSS community owing to its operational simplicity and cost-effectiveness. Thanks to the real-time service (RTS) products of the International GNSS Service (IGS), it is possible to use SF-PPP solutions in real-time tropospheric delay retrieval. Moreover, the advent of newly-emerged satellite systems, such as Galileo and BeiDou, offers significant opportunities to enhance its performance in real-time tropospheric delay estimation. Still, to take advantage of the potential benefits of multi-GNSS combinations, it is required to consider the stochastic properties of observations from different navigation systems. Therefore, the main objective of this study is to improve the tropospheric delay estimation performance of the multi-GNSS SF-PPP solution, including GPS, GLONASS, Galileo, and BeiDou, with advanced stochastic approaches. For this purpose, this study proposes an advanced weighting strategy that employs the variance component estimation method to specify the weights of multi-GNSS observations in the real-time SF-PPP process. The experimental tests reveal that the employment of the advanced weighting strategy in the multi-GNSS solution provides an accuracy of 2.18 cm in tropospheric delay estimation, which means an improvement of 17.4% on average compared with the conventional weighting approach. Besides, separately analyzing the recent performance of each navigation system is another objective of this study. The results indicate that within single-system SF-PPP solutions, Galileo provides the highest accuracy for the zenith total delay (ZTD) estimation with an average RMS error of 3.22 cm which is better than that of GPS solution by 8.3%.