Maximizing SINR for non-orthogonal multiple access with bounded channel uncertainties

Abstract

This letter examines the robust beamforming design multiple-input single-output (MISO) non-orthogonal multiple access (NOMA) downlink systems with imperfect channel state information (ICSI). Sparked by the worst-case performance optimization framework, we consider to maximize the minimum of received signal-to-interference-plus-noise ratios (SINRs) of users, which is not convex with regards of beamforming vectors. The formulated optimization problem is not convex so, to solve the challengeable problem, we first formulate a equivalent optimization problem based on semidefinite programming (SDP). By applying a rank one relaxation and a linear matrix inequality (LMI) the S-Procedure can be used, which leads to exploiting bisection algorithm in order to obtaining the robust optimal beamforming solution. Finally, simulation results demonstrate the proposed design.