User Scheduling and Power Allocation for Precoded Multi-Beam High Throughput Satellite Systems With Individual Quality of Service Constraints

• Published in: IEEE transactions on vehicular technology 2023-01, Vol.72 (1), p.907-923
• Main Authors: Van Chien, Trinh, Lagunas, Eva, Ta, Tung Hai, Chatzinotas, Symeon, Ottersten, Bjorn
• Format: Article
• Language: English
• Subjects: Algorithms; Broadband; Combinatorial analysis; Heuristic methods; High-temperature superconductors; Multi-beam high throughput satellite; Optimization; power allocation; Quality of service; Quality of service architectures; Resource management; Satellite networks; Satellites; Scheduling; sum throughput optimization; Throughput; user scheduling
• ISSN: 0018-9545; 1939-9359
• DOI: 10.1109/TVT.2022.3206264

For extensive coverage areas, multi-beam high throughput satellite (HTS) communication is a promising technology that plays a crucial role in delivering broadband services to many users with diverse Quality of Service (QoS) requirements. This article focuses on multi-beam HTS systems where all beams reuse the same spectrum. In particular, we propose a novel user scheduling and power allocation design capable of providing guarantees in terms of the individual QoS requirements while maximizing the system throughput under a limited power budget. Precoding is employed in the forward link to mitigate mutual interference among the users in multiple-access scenarios over different coherence time intervals. The combinatorial optimization structure from user scheduling requires an extremely high cost to obtain the global optimum even when a reduced number of users fit into a time slot. Therefore, we propose a heuristic algorithm yielding a good trade-off between performance and computational complexity, applicable to a static operation framework of geostationary (GEO) satellite networks. Although the power allocation optimization is signomial programming, non-convex on a standard form, the solution can be lower bounded by the global optimum of a geometric program with a hidden convex structure. A local solution to the joint user scheduling and power allocation problem is consequently obtained by a successive optimization approach. Numerical results demonstrate the effectiveness of our algorithms on GEO satellite networks by providing better QoS satisfaction combined with outstanding overall system throughput.