Joint beamforming and power allocation design in downlink non-orthogonal multiple access systems

We propose a joint design of beamforming and power allocation in a downlink multiple-input multiple-output multiuser system which employs non-orthogonal multiple access (NOMA). We address a new scenario where the users are divided into two groups by their quality of service (QoS) requirements, rather than the location information, such that the users in Group 1 expect to be served with the best efforts whereas the users in Group 2 require to reach target rates. For this scenario, The aim is to maximize the sum rate of the users in Group 1 while guarantee the minimum rates of the users in Group 2. We first apply the semidefinite relaxation (SDR) approach to linearize the quadratic forms of beamforming vectors, and then successively approximate the nonconvex constraints based on the arithmeticgeometric mean inequality to jointly design the beamforming matrices and power allocation. Finally, we show that the proposed algorithm achieves a profound sum rate advantage over the existing ones, and examine the impact of parameters on the convergence rate of the proposed algorithm.