Combining Time-Flexible Satellite Payload with Precoding: The Cluster Hopping Approach

High throughput geostationary (GEO) satellite systems are characterized by a multi-beam wide coverage. However, developing efficient resource management mechanisms to meet the heterogeneous user traffic demands remains an open challenge for satellite operators. Furthermore, the spectrum shortage and the ever increasing demands claim for more aggressive frequency reuse. In this paper, we combine the time-flexible payload capabilities known as beam hopping (BH) with precoding techniques in order to satisfy user traffic requests in areas of high demand (i.e. hot-spot areas). The proposed framework considers a flexible beam-cluster hopping where adjacent beams can be activated if needed, forming clusters with various shapes and sizes. In this context, we present three strategies to design the beam illumination patterns. First, a max-min user demand fairness satisfaction problem; second, a penalty-based optimization is considered to penalize the occurrence of adjacent beams in an attempt to avoid precoding whenever possible. Third, seeking a low-complexity design, we propose a queuing-based approach to solve the problem in a time-slot by time-slot basis trying to provide service to users based on the requested demands. The three methods are discussed in detailed and evaluated via numerical simulations, confirming their effectiveness versus benchmark schemes and identifying the pros and cons of each proposed design