Accuracy of PPP along with the development of GPS, GLONASS and Galileo

In recent years, one can notice a significant development of the PPP, which both increases accuracy and speeds up the convergence time of the receiver position. New or improved computational algorithms have been developed. This development can also be seen in real-time measurements made possible by the IGS RTS. Other new trends are the development of PPP-AR and the use of cheap receivers such as smartphones. However, the development of the PPP method can be particularly seen in multi‑GNSS measurements. This applies to the continuous development of existing GPS and GLONASS systems and the emergence of new Galileo and BDS systems that have a significant impact on PPP. The development of multi GNSS will increase the number of satellites observed, which improves geometry and PDOP, and increases product accuracy or increases the number of available signals and frequencies. The use of multi-GNSS is possible thanks to the IGS MGEX.

This research shows how the accuracy and convergence time by the PPP changes with the development of GPS, GLONASS and Galileo systems. We used the globally distributed MGEX stations for three different weeks, each one from 2017, 2018 and 2019. The analysis was made for different constellations: GPS, GLONAS, Galileo, GPS+GLONAS, GPS+Galileo, GLONASS+Galileo and GPS+GLONASS+Galileo for different cut-off elevation angles: 0⁰, 5⁰, 10⁰, 15⁰, 20⁰, 25⁰, 30⁰, 35⁰ and 40⁰.

Based on the analysis, we show a progressive improvement of accuracy and a shortening of convergence time in recent years. This is especially visible for calculations with multi-GNSS, obtaining the best results for GPS+GLONASS+Galileo for the last analysed period. Already in 2019 on average, about 22 satellites were observed using a total of three systems together. It has also been shown that in 2019, the Galileo system already allows for positioning with high accuracy anywhere on Earth. On average, around 7 Galileo satellites were observed in 2019, where in 2017 on average, fewer than 5 satellites were observed. It has also been shown that the GPS still provides the highest accuracy and has the greatest impact on multi-GNSS positioning accuracy. Even for the GLONASS+Galileo, poorer accuracy was obtained than for GPS‑only. However, for the GLONASS+Galileo solution, a smaller error distribution and lower standard deviation values were obtained than for GPS-only. This may indicate constant bias-related error values (IFB, ISB) and poorer product quality. In addition, for higher elevation angles, it was shown that better accuracy was obtained for Galileo‑only than for GLONASS-only, but only for the third period. It was also noted that for the joint of GLONASS+Galileo, it eliminated errors that occurred in the GLONASS-only, for which, in the second period, much larger errors were obtained than for the other periods. Finally, the influence of multi-GNSS positioning for positioning in constraint conditions was demonstrated by analysing the effect of the elevation angle. It has been shown that even for elevation angle of 40⁰, the use of GPS+GLONASS+Galileo allowed obtaining about 90% of the availability of solutions with accuracy in estimation the position of individual cm.