Remote Sensing, Vol. 11, Pages 571: Seeking Optimal GNSS Radio Occultation Constellations Using Evolutionary Algorithms (Remote Sensing)

Given the great achievements of the Constellation Observing System for Meteorology, Ionosphere, and Climate (COSMIC) mission in providing huge amount of GPS radio occultation (RO) data for weather forecasting, climate research, and ionosphere monitoring, further Global Navigation Satellite System (GNSS) RO missions are being followingly planned. Higher spatial and also temporal sampling rates of RO observations, achievable with higher number of GNSS/receiver satellites or optimization of the Low Earth Orbit (LEO) constellation, are being studied by high number of researches. The objective of this study is to design GNSS RO missions which provide multi-GNSS RO events (ROEs) with the optimal performance over the globe. The navigation signals from GPS, GLONASS, BDS, Galileo, and QZSS are exploited and two constellation patterns, the 2D-lattice flower constellation (2D-LFC) and the 3D-lattice flower constellation (3D-LFC), are used to develop the LEO constellations. To be more specific, two evolutionary algorithms, including the genetic algorithm (GA) and the particle swarm optimization (PSO) algorithm, are used for searching the optimal constellation parameters. The fitness function of the evolutionary algorithms takes into account the spatio-temporal sampling rate. The optimal RO constellations are obtained for which consisting of 6–12 LEO satellites. The optimality of the LEO constellations is evaluated in terms of the number of global ROEs observed during 24 h and the coefficient value of variation (COV) representing the uniformity of the point-to-point distributions of ROEs. It is found that for a certain number of LEO satellites, the PSO algorithm generally performs better than the GA, and the optimal 2D-LFC generally outperforms the optimal 3D-LFC with respect to the uniformity of the spatial and temporal distributions of ROEs.