Multi-GNSS global ionosphere maps comparation based on three stochastic model: Equal weighting processing, single-differenced estimation and VCE

 Jingzhe Shang⁎ , Bofeng Li, Haibo Ge

Abstract 

 The development of Global Navigation Satellite Systems (GNSS) provides great opportunities for deriving Global Ionosphere Maps (GIM) with more data source. The stochastic model of ionospheric observables for different systems is of great importance for global ionospheric modeling with multi-GNSS combination. Variance-Covariance Component Estimation (VCE) is a common method to determine the stochastic model of different GNSS system observations. However, since iterations in VCE is necessary to ensure the con vergence of variance component of different system, it takes a long time in GIM estimations, which affects the efficiency of daily GIM estimations. Thus, improving the efficiency of power ratio estimation is necessary. In this article, a high-efficiency stochastic model, called single-differenced estimation is adopted to determine the power ratios of different system observations, and three models, namely VCE, single-differenced estimation model, and equal-weighting model, are assessed for estimating GIM with multi-GNSS (GPS, GLONASS, BDS, and Galileo) observations. The experimental results show that the performance of VCE and single-differenced estimation model are comparable, and they are better than that of equal-weighting model. With either VCE or single-differenced estimation model, the pre cision of GIM is improved by 0.07 TECu, while the precisions of satellite DCBs are relatively improved by 3.0 %, 2.1 %, 4.8 % and 1.1 % for GPS, GLONASS, Galileo, and BDS, respectively. In addition, difference of Slant TEC (dSTEC) assessment is applied to estimate the precision of GIM. The results show that the precision of GIM generated by single-differenced estimation model (denoted as TJA) and VCE is about 2.74 TECU, which improves by 0.14 TECu with respect to equal-weighing model (denoted as TJE). Moreover, VCE can better estimate the precision of ionospheric observations for each system while the single-differenced estimation has the advantage of computation efficiency.