Multi-GNSS precise point positioning : modeling, ambiguity resolution and quality control

Precise Point Positioning (PPP) aims to provide three-dimensional decimeter-level to centimeter-level positioning information with a single Global Navigation Satellite System (GNSS) receiver in the global scale. Two important issues that improving positioning accuracy and shortening convergence time in PPP have attracted great attention in the past two decades. The integration of multi-GNSS (GPS/GLONASS/BDS/GALILEO) brings more visible satellites and strengthens satellite geometry, making fast PPP initialization possible. In the meanwhile, integer ambiguity resolution enabled PPP, called PPP-RTK, has been booming in the recent decade, and it can help separate satellite biases from the integer ambiguity or compensate these biases in float ambiguities and then improve positioning accuracy with short convergence time. In addition, several models are proposed to solve PPP data processing issues, such as the ionosphere-free model, the uncombined model and the ionosphere-constrained model. Through the development of multi-GNSS, PPP-RTK and PPP models, both above issues are solved to a great extent. However, most of these works focus on one or two parts of these technologies, and only realize the functions without strict quality control. This thesis aims to systemically elaborate, analyze and summarize the modeling, ambiguity resolution and quality control for multi-GNSS PPP from a holistic perspective, then design multi-GNSS PPP-RTK algorithms and a software package to support this research. The main contributions of this thesis are listed as follows: a. We have derived FCB and IRC products using both the IF-PPP and U-PPP models. The theoretical equivalence of FCB estimation based on these two models is justified by comparing respective WL and NL ambiguities. b. We have developed the quality control and reliability analysis procedure for satellite phase bias estimation, and the data snooping method is extended in LS-based FCBand IRC estimation. Both the internal and exter