Abstract
In this paper, we investigate a LEO-satellite-terrestrial network designed to provide high-rate transmission services, on the constraints of limited energy and the stringent timeliness requirements of terrestrial user equipments (UEs). Specifically, we formulate the resource allocation problem for downlink LEO-to-ground transmissions, considering the optimization of user paring, transmit power, and beamwidth, for the sake of jointly minimizing the average Age of Information (AoI) and the average transmission energy consumption. By decoupling the problem, it develops an innovative matching algorithm for user pairing problem, followed by a multi-objective evolutionary algorithm with space division and coverage detection strategy (MOSC) for both optimizing power and beamwidth. These algorithms are executed iteratively with a relatively small number of iterations. Simulation results demonstrate the practical value of the proposed algorithm, showcasing its ability to accurately and efficiently utilize limited communication resources. Consequently, the proposed algorithm can outperform the advanced heuristic algorithms in terms of the convergence and diversity of the obtained non-dominated solutions.