Objectives The continuous improvement and progress of the global navigation satellite system (GNSS) provides a strong foundation for multi-system fusion localization, and with the increase in the number of satellites deployed for multi-system positioning, the probability of gross observational errors is also constantly increasing. Therefore, it is of great significance in satellite positioning to correctly evaluate the quality of the observations and eliminate the errors.
Method Based on reliability theory and the gross error detection method, the minimum detective bias (MDB) and minimum detective error (MDE) are calculated to analyze reliability of multi-GNSS experiment (MGEX)data. Based on these calculation results, we designed and implemented gross error detection experiments.
Results The MDB and MDE values for the dual-frequency solution in fused BDS(BeiDou navigation satellite system)/GPS(global positioning system)positioning were smaller than the valve in the solution from BDS alone, 7.105 m and 22.368 m respectively. Unlike the single-frequency solution, the dual-frequency BDS/GPS combination reduced the MDB by 4.105 m and the MDE by 1.621 m. When adding single gross error to the double-frequency BDS observations, the gross error detection success rate was 99.96% and gross error recognition rate was 90.56%. For the combined system, these values were 99.96% and 99.05% respectively. When adding double gross errors to BDS observations, the gross error detection success rate were 100% and gross error recognition rate was 65.67%, while the gross error detection success for the combined system was 100%, and 71.91% respectively.
Conclusions The MDB and MDE values for multi-system and multi-frequency positioning solutions decline. Furthermore, the gross error recognition rate for the multi-system positioning is higher than a single system alone.