Joint user-beam selection for hybrid beamforming in asynchronously coordinated multi-cell networks

A dense, coordinated deployment of BSs/APs is the most promising approach to address the increasing demand for higher wireless data rates. Using analog beamforming to coordinate and multiplex transmitters while employing digital beamforming locally per transmitter, promises to deliver the performance gains of coordination without high overhead. A central challenge of this hybrid beamforming architecture is how to jointly select the users and beams for each transmitter during a common transmission slot. To address this challenge, we propose a novel greedy user-beam selection scheme, which, similar to user selection algorithms in the context of MU-MIMO, aims to maximize a utility function of the users' rates. Motivated by the luck of formal bounds on such schemes, we propose two additional novel algorithms for which we establish formal performance bounds. The first algorithm is a greedy solution of an associated maximum weight independent set problem with cardinality constraints, while the second is a greedy solution of an associated maximum cardinality problem over a set of feasible links. Simulation results under real world scenarios show that the hybrid beamforming architecture coupled with the proposed user-beam selection algorithms, achieves users rates which are up to 10× higher than those achieved by current uncoordinated deployments.