Performance Optimization for Multi-Gateway NOMA-Beamforming in Multi-Beam SATCOM

To provide high throughput of beamforming (BF) based forward links in multi-beam satellite communication (SATCOM), the state-of-the-art non-orthogonal multiple access (NOMA) technique has been explored, by serving multiple users per beam in the same time slot. Nevertheless, the feeder link limitations and satellite payload constraints need to be considered for BF design and power allocation (PA) optimization in NOMA based transmission. To tackle these challenges, performance optimization algorithms are investigated for applying NOMABF in the upcoming multiple gateway (multi-GW) architectures. Specifically, centralized multi-layer BF is first processed in a central unit with full global channel state information (CSI) and data sharing, which is considered as an upper-bound performance. Next, per-feed available power-constrained BF via maximization of the worst-user signal-to-leakage-and-noise ratio (SLNR) is explored with local CSI for a distributed operation of GWs. Then, weighted sum-rate maximization (WSRM) based PA is applied to optimally use the power resources for more performance increment, where the non-convex WSRM problem is recast into an equivalent weighted sum mean square error minimization (WMMSE) problem for a tractable solution. Simulation results reveal the efficiency of our proposal.