Beamforming Optimization and Power Allocation for User-Centric MIMO-NOMA IoT Networks

Abstract

Ultra high reliability and ultra low latency are the key objectives of the internet of things (IoT) with massive connectivity. To support these objectives, we investigate the resource allocation for the user-centric multi-cell multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) based IoT networks. The macro base station (MBS) equipped with multiple antennas transmits signals to access points (APs) in the backhaul link, and each device can be served by multiple APs in the access link, and the APs serving the same device compose one AP group (APG). The NOMA is applied in each APG to reduce the intra-APG interference. In this paper, the resource allocation problem involving the beamforming optimization and power allocation is formulated as nonconvex optimization problem which is extremely difficult to tackle. In order to reduce the computational complexity, we decompose the resource allocation problem into two subproblems in terms of the beamforming optimization and power allocation. For the beamforming optimization subproblem, the zero-forcing beamforming (ZFBF) algorithm is applied to solve it. When the beamforming strategy is fixed, the power allocation subproblem is still a nonconvex optimization problem. We first transform it as a difference of two convex functions (DC) problem, and then the DC programming method is adopted to optimize it. We further prove that the solution obtained by the DC programming method is one of the local optimal solutions of the original optimization problem. Extensive simulation results are presented to demonstrate the effectiveness of the proposed resource allocation scheme for the user-centric MIMO-NOMA IoT networks.