Station-based SDB correction for GPS triple-frequency observations and its application on relative positioning

Due to the chip shapes of Global Navigation Satellite System (GNSS) live signals differ from ideal conditions, GNSS satellite signals are distorted, which leads to different code biases, namely signal distortion biases (SDB). Existing SDB estimation methods are based on code observation combinations with large noise and face the problem of unreasonable classification, which would seriously affect the reliability of SDB estimation and correction. To solve this problem, a station-based SDB estimation method is proposed in this study. First, a full-rank linear system independent of the ionosphere is derived based on the classical Uofc model. Then, triple-frequency SDBs relative to the reference stations are estimated with three types of virtual observations. To assess and validate the performance of the proposed method, real data of 335 stations from multi-GNSS experiment (MGEX) network in January 2023 are chosen for GPS triple-frequency SDB estimation. The results show that the estimated SDBs of all stations equipped with the same receiver types keep good consistency. In addition, the estimated SDBs are shown to be related to receiver types and antenna types, and maintain high stability over the whole month. Furthermore, the code residuals of 4 zero-baselines from Curtin University CORS and the real-time kinematic (RTK) ambiguity fixing rates of 8 short baselines from the MGEX network are evaluated to validate the correction performance of the estimated SDBs. The results show that the code residuals of zero-baselines are close to zero with SDB correction, and the systematic deviations in the code residuals can be effectively decreased. For the RTK application, the ambiguity fixing rates of the selected short baselines can be increased by about 40.0% after SDB correction. In addition, the percentage of ratio values after ambiguities are fixed with SDB correction is always higher than that without SDB correction, which effectively improves the reliability of RTK ambiguity resolution.