Anomalies in IGS ephemeris and clock products and their influence on the solution of navigation problems

The subject of research of this paper is anomalies in the final products of the International GNSS Service (IGS), namely in the orbits and clock drifts of navigation satellites (NSs). The purpose of research is to de-termine the influence of such anomalies on the accuracy of solving the precise point positioning (PPP) problem. The method of approximation by polynomials of high degrees previously proposed by the authors is used as a method for detecting and distinguishing anomalies in the orbits of navigation satellites. The methodology recommended by the IGS is used in solving the PPP problem. The proposed method for de-tecting and distinguishing anomalies in orbits is applied to the analysis of anomalies in the orbits of GPS navigation satellites. The examples of anomalies that can be detected using the proposed method are dem-onstrated. The brief statistical analysis and comparison of the frequencies of anomalies occurrence in the orbits of GPS navigation satellites published by various IGS analytical centers from 2010 to 2018 are pre-sented. It is shown that orbital anomalies occurring at the boundaries of daily intervals are, as a rule, cor-related with anomalies in clock drifts and have a partially mutually compensating effect on the solution of navigation problems. Experiments showed that when solving the PPP problem, anomalies significantly in-crease the root-mean-square deviation (RMSD) of the solution residual. Two options for solving the prob-lem with anomalous orbits are considered: the exclusion of satellites with anomalous boundaries of daily intervals from the solution and the "correction" of the anomaly in the orbit. The most natural method of correcting orbits (changing the orbit in order to remove large anomalies) at the boundaries of the daily segments of the published final orbits was tested. The exclusion of satellites with anomalies in the orbit turned out to be the most effective from the point of view of PPP problems, since attempts to "correct" the orbit more often led not to a decrease in the RMSD of the pseudorange residuals, but to its increase, which is associated with correlated anomalies in the navigation satellite clock drift. According to the research results, we can conclude: before solving the PPP problems, it is necessary to study the orbits and the navi-gation satellites clocks drifts for the presence of anomalies by the proposed methods and, if possible, to exclude such satellites from the data used to solve the PPP problem. Our proposed methods for detecting and accounting for anomalies in the orbits and clocks of navigation satellites, in addition to obvious appli-cations to solving ground navigation problems, are also applicable to monitoring the quality of the space and ground segments of the GLONASS and GPS systems.