Multi-GNSS real-time precise satellite clock estimation

Nowadays, Global Navigation Satellite Systems (GNSS) play a pivotal role in precise positioning in geosciences and various related domains. Among all GNSS positioning services, real-time precise point positioning (PPP) stands out for its global consistency and flexibility. The quality of real-time services based on PPP hinges particularly on the precision and resilience of satellite orbit and clock corrections. Despite its importance, maintaining continuous and long-term reliable real-time service remains a complex challenge. This dissertation introduces innovative methods, including algorithms, strategies, and system designs, to enhance the performance of real-time services. The core contributions of this work are highlighted below: An innovative approach for real-time clock rapid recovery is developed. Traditional clock estimation strategies require extensive time, often hours, to regain accuracy after interruptions, leading to significant gaps in real-time products and deteriorating the performance of real-time PPP services. The proposed method accelerates clock estimation restarts by leveraging historical data and previously saved filter state information, thus minimizing downtime. Additionally, it employs multi-thread processing to address the issue of thread concurrency effectively. A novel Decentralized Clock Estimation (DECE) strategy for high-rate real-time multi-GNSS clock estimation is also proposed. The complexity of updating clocks in real-time at high frequencies poses significant challenges, especially with the expansion of multi-GNSS constellations and the increasing number of stations. The DECE strategy enables multiple Un-Differenced (UD) estimators and Epoch-Differenced (ED) estimators running on separate computers and even locations, with their outputs merged to produce a unified clock estimation to reduce the computational burden and to improve the product availability. The experimental results show that the DECE method greatly enhances clock estimation efficiency while maintaining performance nearly identical to that of the original clock products. The evolution of the GNSS Real-Time Precision Positioning Service (RTPPS) system at the German Research Centre for Geosciences (GFZ) is comprehensively outlined in this dissertation. The system integrates real-time orbit determination, clock estimation, Uncalibrated Phase Delay (UPD) estimation, and PPP functionalities along with regional augmentation features, with a particular emphasis on system architecture and performance evaluation. Additionally, it extensively analyzes the impact of network configuration on real-time clock estimation. The Real-Time Positioning and Monitoring (RTPM) system is developed to retrieve real-time high-precision ground displacement for hazard monitoring and early warning in the framework of the Early-Warning and Rapid ImpaCt Assessment with real-time GNSS in the Mediterranean (EWRICA) project. The RTPM system has currently been deployed to the Rete INtegrata Nazionale GNSS (RING) network in Italy and the Badan Informasi Geospasial (BIG) network in Indonesia. its exceptional positioning performance has been demonstrated operationally. In summary, this dissertation primarily focuses on improving the performance of real-time precise positioning service, particularly in the real-time satellite clock estimation. The introduction of various new strategies has significantly improved the efficiency and stability of real-time GNSS service. Both experimental validation and in-field operational deployment through the EWRICA project have demonstrated that the improved RTPPS system can offer users a variety of PPP services, yielding promising results.