Joint power and antenna selection optimization for energy-efficient large distributed MIMO networks

Large multiple-input multiple-output (MIMO) network promises high energy efficiency using a large number of antennas. To reduce the signaling overhead of obtaining the full channel state information, we propose a downlink antenna selection scheme for large distributed MIMO networks with regularized zero-forcing (RZF) precoding. We study the joint optimization of antenna selection, regularization factor, and power allocation to maximize the average weighted sum-rate. The problem is non-trivial due to its combinatorial and non-convex nature. We decompose the problem into subproblems, each of which is solved by an efficient algorithm. For very large distributed MIMO networks, we obtain a capacity scaling law and show that there is an asymptotic decoupling effect, which can be exploited to simplify algorithms and physical layer processing. Simulations show that the proposed scheme achieves significant gain over the baseline.