An Optimization Method for the Gateway Station Deployment in LEO Satellite Systems

Abstract

Low Earth Orbit (LEO) satellite networks play a major role to provide communication support for the regions beyond the coverage of terrestrial network systems. The positions of gateway stations have influence on the time delay, power consumption and throughput of LEO satellite networks, making it important to deploy the gateway stations at the best position. Aiming at minimizing the inter-satellite hop count, we are the first to study the deployment problem of gateway stations. The gateway deployment problem is formulated as a non-convex optimization problem. According to the characteristics of the deployment problem, we propose an improved genetic algorithm (GA) and an improved simulated annealing algorithm (SA) not only to get the optimal deployment location, but also to greatly reduce the computational complexity. Simulation results and analyses show that compared with random deployment, the optimal locations of the gateway stations obtained by the proposed algorithms can decrease 42.6% average hop count of inter-satellite links and increase 4.4% average capacity of feeder links.