Abstract: Objectives: The Galileo High Accuracy Service (HAS) transmits precise orbit, clock and code bias products of the Galileo and GPS dual systems to users worldwide for free through satellite-based E6b signal or the Internet, offering a new approach for global real-time Precise Point Positioning (PPP). Due to the fact that Galileo has been formally providing HAS for a relatively brief period, there is still little research about exploring service consistency on a global scale and PPP performance in real dynamic environments. Methods: The decoding method of HAS message, the usage of corrections and the PPP process were explored and described. The availability and accuracy of orbit and clock offset were comprehensively analyzed using satellite-broadcast HAS data. Thirteen Multi-GNSS Experiment (MGEX) stations in different regions were selected for comparison and analysis of HAS PPP performance in static and simulated kinematic modes. Additionally, a real dynamic test of HAS PPP was conducted in the Wuhan region for a more comprehensive evaluation of positioning performance. Results: The results show that the Galileo and GPS satellite corrections have high average availabilities of 98.9% and 94.4% respectively. Compared with the Final satellite products of Wuhan University (WHU), the one-dimensional root mean square (1D RMS) errors of the Galileo and GPS satellite orbit products are 6.2cm and 6.6cm, respectively.Additionally, the standard deviation (STD) of the clock errors are reported as 0.14ns for Galileo and 0.23ns for GPS, respectively. Those indicate a significant enhancement in accuracy compared to the broadcast ephemeris. Experimental results demonstrate that the HAS PPP solution achieves centimeter-level positioning accuracy in static scenarios. In simulated kinematic mode, HAS PPP maintains horizontal positioning precision at the centimeter level while exhibiting vertical accuracy ranging from 1 to 2 decimeters. The positioning accuracy of real dynamic mode can achieve 1~4 decimeter level. GPS error is slightly higher, but can still meet sub-meter positioning requirements of HAS. The convergence time and elevation errors of the positioning initial phase are influenced significantly by the region distribution of stations. Conclusions: Overall, thanks to the high accuracy of the orbit and clock corrections, HAS can effectively meet the demand for real-time precise positioning. The HAS PPP accuracy and convergence performance will be further improved in the future with the development of HAS Phase 2.