Estimation and analysis of multi-GNSS observable-specific code biases

Abstract

The proper handling of code biases is essential for realizing precise ionospheric modeling, positioning and timing. It is common to treat code biases in a differential mode as a differential code bias (DCB). With the modernization of GPS and GLONASS and the implementation of Galileo and BDS, the traditional DCB calibrations are complex and not easily extendable to different frequencies and modulations. An alternative treatment of code biases is to use observable-specific signal biases (OSBs). In this contribution, all possible OSBs are estimated for the latest GNSS and analyzed from the perspectives of precision, consistency and stability. The precisions of the GPS and Galileo OSBs are significantly better than those of the GLONASS and BDS OSBs. Considering the inter-frequency bias (IFB) of GLONASS and the inter-system bias (ISB) of BDS can improve the precisions of their OSB estimates. OSB comparisons among different agencies reveal that GPS and Galileo show good agreement at the level of 0.2–0.3 ns, while the differences of the GLONASS and BDS OSBs reach 0.5–1.0 ns. In addition, agreement of 0.4–0.5 ns is demonstrated for IGSO and MEO OSBs, while the consistency of GEO OSBs is worse by a factor of 2–3. The stability of the OSB estimates is at the level of 0.03–0.09 ns for GPS, 0.10–0.25 ns for Galileo, 0.14–0.48 ns for GLONASS, and 0.16–0.44 ns for BDS. In general, the BDS-3 OSB estimates show better stability than the BDS-2 OSBs. Moreover, the code biases at the same or at a close central frequency show similar performance. This is particularly obvious for Galileo and BDS, which adopt the dual-frequency constant envelope multiplexing (DCEM) technique. For instance, the code bias estimates of C5Q, C5X, C7Q, and C8Q are close to each other for individual Galileo satellites, and the BDS code biases of C5P, C7Z, and C8X are comparable to each other.