Resource Management Optimization for Interference-Limited Multi-User Satellite Networks (ROIMUS)

点击翻译按钮获取中文摘要

Future satellite communications have two major properties. First, the number of feeds, i.e., the number of antenna elements, at the satellite are in the order of several hundreds. Additionally, transmission is possible in time slots on a large number of carriers. Second, static and mobile users must be served, whose channels have the properties of a line-of-sight connection, i.e., the channel covariance matrices are rank one.For such satellite systems, the joint design of beamforming and user scheduling is considered that is able to deal with static and mobile users based on a robust problem formulation. Globally optimal solutions are derived based on a dual decomposition approach and monotonicity properties of the objective. Due to the large dimensions, i.e., many antennas, carriers, and users, heuristics must be employed to find low-complexity solutions. On the other hand, the large dimensionality of the problem allows for a large system analysis.Besides the adaptive beamforming and the state-of-the-art fixed beamforming with scheduling, a discretely adaptive beamforming scheme is investigated where the ground station can choose over a set of discrete beamforming patterns. Moreover, the case of a satellite cluster is considered besides the conventional single satellite setup. The additional difficulties of the design for a satellite cluster, where the users are served by several satellites, are the separate power constraints and the structural constraints of the beamformer. Additionally, different cooperation levels of the ground stations controlling the satellites are investigated.