Sum-rate Maximization in NOMA-based mmWave Analog Beamforming under Imperfect CSI

In this paper, we investigate the problem of sum-rate maximization in a millimeter-wave (mmWave) multi-user downlink system employing non-orthogonal multiple access (NOMA). Specifically, we consider the design of optimal transmit beamformer and power allocation that is robust in the presence of channel state information (CSI) errors. In order to reduce hardware complexity, we consider analog beamforming using a single RF chain and a double-phase shifter network. The original formulation of this problem is nonconvex; however, we reformulate this problem as an equivalent but tractable weighted mean squared error (MSE) minimization problem. We propose an iterative algorithm to solve this reformulated problem. We demonstrate the robust performance of the proposed algorithms via simulations.