Coordinated multicast beamforming for multi-cell massive MIMO systems
In this thesis, we consider the efficient coordinated multi-cell multicast beamforming design in massive massive multiple-input multiple output (MIMO) cellular networks. For the problem that aims to minimize the base station transmit power, we obtain the optimal multicast beamforming solution structure. Utilizing the optimal structure to reduce the computational complexity, we propose to determine the parameters in the optimal beamforming solution via semi-definite relaxtion (SDR) or successive concex approximation (SCA). To further design a fast algorithm applicable to massive MIMO systems, we apply the alternating direction method of multipliers (ADMM) technique and propose an ADMM-based first-order algorithm to solve the quality of service (QoS) problem, which decompose the QoS problem into subproblems with closed/semi-closed form updates. Following this, we consider the max-min fair problem that aims to maximize the minimum signal to the interference and noise (SINR) of each user. We present the optimal max-min fair (MMF) coordinated multicast beamforming solution structure through the inverse relation between the QoS and MMF problem. We propose to use the projected subgradient algorithm (PSA) to obtain a solution to the MMF problem, which is a fast algorithm with closed-form updates and low computational complexity. Simulation results show that our proposed methods can achieve near-optimal performance. Besides, with much lower computational complexity in a massive MIMO system than existing algorithms.