Resource Allocation for MC-DS-CDMA in Beam-Hopping LEO Satellite Networks

Multicarrier direct-sequence code division multiple access (MC-DS-CDMA) is applicable to future low Earth orbit (LEO) satellite systems due to its superiority in soft beam, spectral, and power efficiency, as well as its robustness to interference and multipath fading. Since the ground traffic demand of LEO satellite systems features time-varying characteristics and nonuniform spatial distribution, it is challenging to support users efficiently and flexibly with limited beam, frequency, and power resources. To this end, this article investigates a resource allocation scheme minimizing the energy consumption together with the unmet system capacity (USC) of the downlink MC-DS-CDMA beam-hopping LEO satellite systems. Specifically, we first formulate the problem as a mixed integer multiobjective nonconvex optimization problem, which is generally intractable. To solve this complicated problem, we then propose a multidomain decoupled allocation algorithm based on relaxation and convex approximation inner–outer loop (RCAIL). With the aid of successive convex approximation, variable relaxation, and big-M methods, we can obtain the suboptimal solutions of the original optimization problem with a polynomial time computational complexity. The scheduling strategies of beam, subcarrier, code, and transmission power can be thereby derived. The simulation results illustrate that the proposed multidomain decoupled allocation algorithm can significantly improve USC performance with lower total energy consumption compared with conventional allocation strategies. Numerically, the USC of the proposed algorithm is 38$\%$ lower than that of the conventional algorithm when the energy consumption is consistent. Moreover, the average delay of the proposed algorithm is also superior to the other beam-hopping strategies.