Global optimization of GNSS station reference networks

The IGS08 reference network, adopted by the International GNSS Service (IGS) in 2011, and IGS08 core network are both of primary importance for the international GNSS community. They were empirically designed so as to fulfill several standards, such as the quality of the ITRF2008 positions and velocities of the involved stations, the reliability of the data they provide, and their global distribution. The present study shows that such global reference networks and well-distributed sub-networks can be objectively and rigorously designed with genetic algorithms (global optimization stochastic algorithms), on the basis of equivalent criteria. The IGS08-like reference networks and IGS08 core-like sub-networks provided by genetic algorithms respectively outperform the IGS08 and IGS08 core networks for the specific criteria in play. This opens the way to future essential applications: objective design of the next IGS reference and core networks, construction of dynamic core networks, and computation of user-specific global or regional GNSS reference networks.