Regional Ionospheric Modeling with Raw Observations and Its Application in BDS Single-Frequency PPP

The single-frequency precise point positioning (SFPPP) method has attracted widespread attention for its high precision and low cost, while the time delay caused by the ionosphere is one of the most problematic error sources in its data processing. With the completion of the BeiDou Global Navigation Satellite System (BDS), the users can access BDS positioning, navigation, and timing (PNT) service around the world. It is crucial to generate high-precision ionospheric products to improve the positioning accuracy and accelerate the convergence. However, the existing Global Ionospheric Map (GIM) products cannot satisfy the requirements of regional users for precise positioning, due to its poor spatial and temporal resolution. In this contribution, we estimate the biased slant ionospheric delay (SID) based on GPS dual-frequency (DF) raw observations from 30 stations in Australia and develop the regional ionospheric model (RIM). The performance of SFPPP with ionosphere corrected by RIM is evaluated using the BDS-2/BDS-3 observations from several stations and compared with the corresponding results obtained by CODE-GIM. Accounting for the long convergence time of uncombined (UC) SFPPP, we constrain the ionospheric delay parameter with RIM. And a time-varying weight constraint is attached to weaken the effect of RIM on the positioning accuracy after convergence. The results show that the RMS of the VTEC difference between RIM and CODE-GIM is better than 2 TECU. The RMS of BDS SFPPP with ionosphere corrected by RIM is better than 0.4 m and 0.3 m in north and east directions, respectively, and the average vertical accuracy is 0.53 m. The ionosphere-weighted (IW) SFPPP with RIM can significantly improve the convergence speed compared to the UC-SFPPP.