Low-Complexity Beam Selection Algorithms for Millimeter Wave Beamspace MIMO Systems

Abstract

We consider the downlink multi-user multiple-input multiple-output (MIMO) systems, operating at millimeter-wave frequencies and equipped with a large number of antenna elements at the access point. Beamspace MIMO with beam selection offers an attractive solution for reducing the number of required radio frequency chains in such systems. We propose two beam selection algorithms. The first one is a heuristic greedy algorithm that enjoys low computational complexity with moderate performance. By modeling beam selection as maximum weight matching over a bipartite graph, we propose beam selection based on the Kuhn-Munkres algorithm and, importantly, introduce a new formal framework for beam selection.