Comparative analysis of global ionospheric models used in GNSS data processing based on selected stations

Among the many error sources affecting GNSS (Global Navigation Satellite System) positioning accuracy, the ionosphere is the cause of those of the greatest value. The ionized gas layer, where also free electrons are present, extends from the upper atmosphere to 1,000 km above the Earth's surface (conventionally). As the GNSS satellite orbits altitude is more than 20,000 km, the wave transmitted from the satellite to the receiver on the Earth’s ground passes through this layer, but not unscathed. The ionosphere is a dispersive medium for the electromagnetic waves in the microwave band, including UHF (Ultra High Frequency) waves transmitted by GNSS satellites. As a result, the group velocity of the wave decreases, while its phase velocity – increases.

Ionospheric delay compensation methods include among others multi-frequency measurements;  however, when considering measurements on one frequency, the usage of ionospheric models is an option. The key element is the number of free electrons, its inclusion in the course of calculations is possible thanks to the TEC (Total Electron Content) maps. Taking into account the variability of the coefficient in the daily and annual course, as well as depending on the activity of the Sun and its 11-year cycle, it is important to use the current value for a given place and time.

For the European Galileo satellite system a dedicated ionospheric model NeQuick-G was developed. As a simple modification of the formula allows it to be applied to other satellite systems, it can be considered in a broader context, regardless of the system and receiver location. In our study the TEC maps published by IGS are used as the comparative data. As a reference, the station located in Warsaw, Poland, is adopted.

The subject of this research is the reliability and validity of the model in equatorial region. The analysis is performed for the stations belonging to the IGS (International GNSS Service) network, located in the discussed area. For each hour of the day, independently for each month of 2019, statistic parameters are determined for both models as well as for the difference between them. The results are analysed taking into account the local time of individual stations. The decisive element is the comparison of the station position time series during disturbed and quiet ionospheric conditions (selected based on the K-index), using each of the models (single-frequency observations). The station coordinates are determined from GPS (Global Positioning System) data using the PPP (Precise Point Positioning) method; the position determined for the iono-free combination (dual-frequency observations) is used as a reference.

The ionospheric delay is directly proportional to the value of the TEC parameter. The difference between the models, exceeding on average even 20 TECU (Total Electron Content Unit) during some periods, translates into a station coordinate differences. The presented analysis may indicate the need for local improvement of global ionospheric models in the discussed region, which will consequently affect the GNSS positioning quality.