An Evaluation of Web-based GNSS Processing Services in Saudi Arabia
Abstract
Several relative and precise point positioning (PPP) programs are available online for static
GNSS data processing. These programs are well established and developed by several
organisations worldwide. PPP-based services include the Automatic Precise Positioning
Service (APPS), the Canadian Spatial Reference System Precise Point Positioning service
(CSRS-PPP), GPS Analysis and Positioning Software (GAPS), Trimble RTX and magicGNSS.
Software based on a relative GNSS approach includes the Online Positioning User Service
(OPUS), the Australian Surveying and Land Information Group Online GPS Processing
Service (AUSPOS) and the Scripps Coordinate Update Tool (SCOUT). These web-based
engines can process data from around the globe to provide precise positioning solutions, which
may help minimise personnel, equipment and logistics expenses compared with traditional
GNSS surveying or when establishing reference stations in remote areas (Ocalan, 2015).
In order to see how useful these solutions are in Saudi Arabia, data sets from six continuously
operating reference stations (CORS) within Saudi Arabia were processed through the web-
based software to evaluate several metrics. First, the accuracy of positional solutions for two-
hour, random-day data after the midnight session was assessed on all online services, and the
models used and utilities provided by each solution were investigated. Second, the effects of
observations made during different atmospheric conditions (ionospheric and tropospheric)
were examined using the APPS and Trimble. Measurements were taken at various times
(08:00–10:00, 11:00–13:00 and 14:00–16:00) on four different days throughout the seasons to
cover different weather conditions. The four days were day 251, 2017; day 005, 2018; day 094,
2018; and day 181, 2018. Finally, the accuracy of processing the data was evaluated at different
data lengths (one hour, two hours, four hours, six hours and twenty-four hours) using APPS
and Trimble.
The results from each service were compared with the official coordinates obtained from the
Saudi General Commission for Survey (GCS) using a root-mean-square error (RMSE)
equation. Solutions based on a PPP approach showed more accurate positioning than those
based on relative techniques. Specifically, APPS and Trimble performed very well compared
to other systems in terms of positional accuracy, models used and utilities provided. For the
results of using both services on different days throughout the seasons, the most accurate
positioning was obtained in the morning session of day 094, 2018, as well as the afternoon and late afternoon sessions of day 181, 2018. Increasing the observation spans from one to twenty-
four hours did not significantly minimise the RMSE of the differences. However, two-hour
data lengths have important merits indicating the lowest RMSEs compared to other sessions.