Joint Beamforming and Power-Splitting Design for Cooperative Non-Orthogonal Multicast

We propose a cooperative non-orthogonal multicast scheme for multiple-input single-output (MISO) systems, where the transmitter sends the superimposed signal to two multicast groups. After successfully detecting all signals, the cell-center users (CCUs) in one multicast group help relay signal to the cell-edge users (CEUs) in another group to enhance signal reception. Simultaneous wireless information and power transfer is adopted at the CCUs to assist information relaying. In such a system, to minimize the transmission power of the system while satisfying the quality of service requirements of all the users, an optimization problem is formulated to jointly design the transmitter beamformers, the distributed CCUs beamformer as well as the power-splitting ratios. To solve the non-convex problem, we first equivalently transform it into a difference of convex (DC) programming. Then a low-complexity iterative algorithm based on the constrained concave convex procedure (CCCP) is proposed to solve the DC programming one. In addition, a robust joint beamforming and power-splitting scheme is proposed by assuming imperfect channel state information (CSI). The infinite constraints caused by CSI uncertainties are converted into finite ones utilizing S-procedure. To obtain a rank-one locally optimal solution, the penalty function method and CCCP algorithm are adopted. Simulation results reveal that the proposed cooperative scheme can greatly reduce the transmission power and outperform the baselines within a certain range. Moreover, the robustness and effectiveness of the robust design under imperfect CSI case have been validated.