Cross-layer optimization for satellite-terrestrial heterogeneous networks

The advanced multibeam satellite equipped with phased array antenna can effectively serve a large number of users over its coverage area by beamforming narrow spotbeams and managing interbeam interference. In this paper, we in-vestigate a cross-layer scheme of resource allocation and user scheduling when the multibeam satellite has a choice between routing signals to a ground gateway/feeder antenna/gap filler or sending directly to end user terminals. The satellite scheduler faces a problem of choosing a better signal path by considering transmission diversity to improve reliability and throughput. We solve a general utility problem based on channel conditions and potential interbeam interference that is primarily determined by geographical distribution of users and gateway locations. The optimum solution chooses users to be served at each transmit opportunity and the optimum transmission strategy that gives the highest marginal return of the utility in terms of transmit power allocation and signal path selection. Our cross-layer optimized scheme for satellite-terrestrial heterogeneous networks takes into joint account the PHY layer beamforming, the MAC layer scheduling, and the Network layer routing path selection. From the optimum scheduling policy, a computationally efficient algorithm is proposed. Simulation results show that the use of gateway/feeder antenna can increase the sum capacity of the multibeam satellite regardless of the interference level.