QoS Guaranteed Load Balancing in Broadband Multi-Beam Satellite Networks.

Aiming at the load balancing problem for uneven traffic demands among different beams in a broadband multi-beam satellite network (BMSN), we present a mathematical analysis and propose a load balancing algorithm to improve the efficiency of system resource allocation. In contrast to existing studies, our research focuses on the heterogeneous real-time services, including both VoIP and MPEG traffic flows. Subject to the delay constraint of real-time traffic, we introduce a QoS guaranteed resource scheduling framework for each beam. Based on this framework, we simulate the load imbalance scenario through the Poisson random point process. Then, we present joint adjacent beams and hopping beams load balancing algorithm to maintain the better match between the system resource and uneven traffic demands for BMSN over the satellite lifetime. The key point of our algorithm is intended to reduce the packet loss ratio (PLR) for each user and improve the BMSN's throughput. Simulation results show that the proposed load balancing algorithm can effectively reduce the average packet loss ratio of all beams by 71.25% and improve the average throughput of all beams by 13.97%.