A fast GNSS satellite selection algorithm for continuous real-time positioning

Benefiting from global navigation satellite systems (GNSS), the spatial distribution of satellites and the positioning accuracy of receivers have improved. However, the processing load for tracking all visible satellites has increased significantly. In this study, the temporal correlation in satellite selection results is analyzed, and a fast satellite selection algorithm based on a modified beetle antennae search (MBAS) is proposed to fulfill the requirements of continuous real-time positioning. This approach encodes the satellite, regards the satellite selection set as the position of the beetle, and generates beetle antennae signals through single- and multi-direction searches to randomly optimize the selected satellites. In addition, the geometric dilution of precision is used as an adaptive function to evaluate the intensity of the antennae signal, and the position of the beetle is updated to gradually approach the optimal solution. Experimental results show that the application of MBAS provides better positioning accuracy, has stronger time correlation, and derives in lower computational complexity than other meta-heuristic algorithms, such as the Genetic Algorithm and Particle Swarm Optimization. The proposed algorithm can be applied to continuous and real-time multi-GNSS positioning with different number of satellites.