Calculation of GNSS differential code biases via ionospheric delay interpolation

In this study, the station bias of a stationary Global Navigation Satellite System (GNSS) station located on the rooftop of the Zurich University of Applied Sciences (ZHAW) building was calculated and analyzed. Using a Septentrio mosaic-X5 development board receiver, data was collected over a four-week period in early 2024. The analysis focused on calculating station biases for GPS L1-L5 and Galileo E1-E5a frequency combinations. GNSS signals experience frequency-dependent ionospheric delays, which must be accounted for when calculating station biases. Since station biases and ionospheric delays are indistinguishable in the measurements, the ionospheric delay must first be calculated to isolate the station bias. To achieve this, a methodology based on pseudorange differences was applied, leveraging surrounding International GNSS Service (IGS) stations to interpolate ionospheric delays and isolate the station-dependent bias term. To validate the methodology, station biases of the IGS station in Hügelheim, Germany (BRMG) were computed and compared with corresponding values taken out of published Differential Code Bias (DCB) product files. A comparison of the mean, standard deviation and range between the 2.5% and 97.5% quantiles showed high levels of commonality in variance and statistical distribution, as confirmed by Levene tests. For the ZHAW station, calculated biases were compared to BRMG and three other IGS stations. While biases for station BRMG and the station in Padova, Italy (PADO) were statistically similar, stations Oberpfaffenhofen, Germany (OBE4) and Zimmerwald, Switzerland (ZIM2) showed greater variability. Despite the uncertainties in comparing the self-calculated station bias to the published DCBs, the statistical similarity with IGS stations like BRMG and PADO indicate a robust methodology. Therefore, the results for ZHAW can be considered as reliable values. The analysis of ZHAW station biases revealed mean values of 1.95 m for GPS and 1.51 m for Galileo, calculated over the whole evaluation period. These values can be used as station bias corrections for a limited period, assuming relative stability over time. However, to ensure long-term accuracy, future work should focus on extending the observation period to assess potential seasonal variations. If significant variations are observed, seasonal station bias values could serve as an intermediate solution, providing more precise corrections without requiring daily computations.