Statistical Comparison on Accuracies of Web-Based Online PPP Services

The precise point positioning technique (PPP), an absolute positioning method, is widely used in geodetic point positioning. This study investigates the accuracy of the technique through statistical comparisons of the root mean square errors (RMSE), which are calculated with coordinates obtained from online PPP services. For this purpose, we carried out a two-stage investigation. First, we chose two days, a high geomagnetic activity day and a quiet one. During the high geomagnetic activity day, the Kp index went above 7, and the Dst was below $-170\mathrm{nt}$. We uncovered the effect of geomagnetic activity by using the Automatic Precise Positioning Service (APPS), Canadian Spatial Reference System Precise Point Positioning (CSRS-PPP), and magic Global Navigation Satellite System (magicGNSS) PPP services. The results show that the three-dimensional (3D) RMSE are $\sim 1.5$ times higher during the high geomagnetic activity day during 1-, 2-, 4-, and 6-h session durations, than in the quiet day. However, when the session duration increases to 24 h, the effects of geomagnetic activities are significantly eliminated. In the second stage, we chose 31 consecutive days without significant geomagnetic activities and obtained PPP-derived coordinates for 1-, 2-, 4-, 6-, and 24-h from APPS, CSRS-PPP, magicGNSS, and Trimble real-time extended (RTX). We compared the PPP-derived coordinates with Australian Online GPS Processing Service (AUSPOS)- and Online Positioning User Service (OPUS)-derived reference coordinates. As a result, we determined that, as the session duration increases, the 3D RMSE decreases, and therefore the position accuracy increases. In terms of 3D RMSE, CSRS-PPP yielded the best results in all scenarios. Finally, we concluded that 3D RMSE decreased by approximately 55.8%, 18.1%, 7.7%, and 6.2%, when session durations increased from 1 h to 2 h, 2 to 4 h, 4 to 6 h, and 6 to 24 h, respectively.