Networking with multi-service GEO satellites: cross-layer approaches for bandwidth allocation

Cross-layer Radio Resource Management (CL-RRM) has been recently investigated by quite a few research groups in wireless communications. In the specific satellite-networking environment, the paper presents an overview of different CL-RRM techniques devoted to dynamic bandwidth allocation, whose interactions span the physical, data link, network and transport layers, in various combinations. A multiservice setting is considered, in the presence of variations in both traffic and channel conditions. Regarding the latter, bit and coding rate adaptation are adopted as fade countermeasure, and their effect on the higher layers is modelled as a bandwidth reduction. Traffic models and methodologies for dynamic bandwidth allocation and performance optimization are discussed. Numerical examples are presented to highlight throughput/fairness trade-offs for long-lived TCP connections that share multiple channels with different fading depth.