Evaluation of the recent performance of analysis-center-specific OSB products for PPP-AR: a focus on availability, convergence time and positioning accuracy

Abstract Precise point positioning (PPP) is a widely adopted absolute GNSS positioning technique known for its high accuracy. However, it suffers from long convergence times, which is its main limitation. To address this, PPP with ambiguity resolution (PPP-AR) methods have emerged as a promising approach to enhance both convergence and positioning accuracy. In recent years, four International GNSS Service (IGS) analysis centers—Center for Orbit Determination in Europe (CODE), Centre National d’Etudes Spatiales, German Research Centre for Geosciences, and Wuhan University Multi-GNSS Experiment—have begun to provide observable-specific signal bias (OSB) products, which have gained considerable attention for the facilitation of PPP-AR implementation. This study evaluates the recent performance of these products throughout the year 2024 using the open-source multi-GNSS PPP software, PPPH, which we adapted for PPP-AR solutions. First, data availability was assessed by computing the annual availability percentage for each product and the average number of available OSB values for dual-frequency observations; short- and long-term OSB stability was also analyzed. Performance evaluation was then conducted using 24 h datasets from 20 global IGS stations. According to the results, PPP-AR is more favorable for GPS and Galileo than for BeiDou in single-constellation configurations. Moreover, the results indicate that OSB product selection can significantly influence convergence behavior, depending on the constellation used, while the contribution of PPP-AR on positioning accuracy after the convergence period is limited, regardless of the choice of OSB product. Notably, CODE stands out among the analysis centers with its superior performance and data availability. When used in PPP-AR, CODE’s OSB products yield the shortest average convergence times—12.5 min for GPS-only, 14.4 min for Galileo-only, and 7.8 min for GPS + Galileo—to achieve decimeter-level 3D positioning accuracy.