User Scheduling for Coordinated Dual Satellite Systems with Linear Precoding

TL;DR

This paper introduces a low-complexity, heuristic user scheduling algorithm for dual satellite systems with linear precoding, significantly improving spectral efficiency by reducing inter-user and inter-satellite interference through partial cooperation.

Contribution

It proposes a novel iterative user allocation algorithm that enhances spectral efficiency in dual satellite systems by coordinating user sets to mitigate interference.

Findings

Significant increase in spectral efficiency compared to conventional methods.

Improved zero forcing performance through optimized user orthogonality.

Effective reduction of inter-satellite interference with moderate complexity.

Abstract

The constantly increasing demand for interactive broadband satellite communications is driving current research to explore novel system architectures that reuse frequency in a more aggressive manner. To this end, the topic of dual satellite systems, in which satellites share spatial (i.e. same coverage area) and spectral (i.e. full frequency reuse) degrees of freedom is introduced. In each multibeam satellite, multiuser interferences are mitigated by employing zero forcing precoding with realistic per antenna power constraints. However, the two sets of users that the transmitters are separately serving, interfere. The present contribution, proposes the partial cooperation, namely coordination between the two coexisting transmitters in order to reduce interferences and enhance the performance of the whole system, while maintaining moderate system complexity. In this direction, a heuristic, iterative, low complexity algorithm that allocates users in the two interfering sets is proposed. This novel algorithm, improves the performance of each satellite and of the overall system, simultaneously. The first is achieved by maximizing the orthogonality between users allocated in the same set, hence optimizing the zero forcing performance, whilst the second by minimizing the level of interferences between the two sets. Simulation results show that the proposed method, compared to conventional techniques, significantly increases spectral efficiency.