Resource allocation is a fundamental task in cell-free (CF) massive multi-input multi-output (MIMO) systems, which can effectively improve the network performance. In this paper, we study the downlink of CF MIMO networks with network clustering and linear precoding, and develop a sequential multiuser scheduling and power allocation scheme. In particular, we present a multiuser scheduling algorithm based on greedy techniques and a gradient ascent {(GA)} power allocation algorithm for sum-rate maximization when imperfect channel state information (CSI) is considered. Numerical results show the superiority of the proposed sequential scheduling and power allocation scheme and algorithms to existing approaches while reducing the computational complexity and the signaling load.

翻译：资源分配是无蜂窝（cell-free, CF）大规模多输入多输出（massive multi-input multi-output, MIMO）系统中的一项基本任务，可有效提升网络性能。本文研究了采用网络聚类与线性预编码的CF MIMO网络下行链路，并提出了一种序贯多用户调度与功率分配方案。具体而言，针对不完美信道状态信息（channel state information, CSI）场景，为最大化系统和速率，我们分别提出了基于贪婪技术的多用户调度算法和基于梯度上升（gradient ascent, GA）的功率分配算法。数值结果表明，相较于现有方法，所提出的序贯调度与功率分配方案及算法在降低计算复杂度和信令开销的同时，展现出优越性。