Resource management in multi-cell multiple-input multiple-output non-orthogonal multiple access (MIMO-NOMA) is challenged by computational complexity, flexible clustering, and potential channel correlation. In this paper, we focus on a combined resource allocation problem: NOMA mobile user (MU) clustering and the base station (BS) selection, to improve system data rate. Different from sum data rate maximization and max-min fairness, we introduce a new objective function, i.e., relative fairness, which integrates MU fairness into system data rate optimization to overcome the domination effect of BS in advantaged situations of sum data rate improving. Moreover, we derive the closed form solution of MIMO-NOMA resource allocation for a single cluster, and it can be employed for any size of cluster. Furthermore, we propose a new two-side coalitional matching approach to jointly optimize MIMO-NOMA clustering and BS selection, which is able to balance the tradeoff between MUs’ individual benefits and the overall network performance. The proposed approach is core stable. Pauta-criterion is employed on system performance evaluation to provide a judgement on win-win solutions. In simulation, extensive comparisons provide insightful understanding of our proposed MIMO-NOMA clustering strategy, relative fairness, and the proposed two-side coalitional matching approach.

中文翻译：

---

多小区 MIMO-NOMA 系统中联合 MIMO-NOMA 聚类和基站选择的两侧联合匹配方法

多小区多输入多输出非正交多址 (MIMO-NOMA) 中的资源管理受到计算复杂性、灵活聚类和潜在信道相关性的挑战。在本文中，我们关注组合资源分配问题：NOMA 移动用户 (MU) 聚类和基站 (BS) 选择，以提高系统数据速率。不同于和数据速率最大化和最大最小公平性，我们引入了一个新的目标函数，即相对公平性，它将MU公平性融入系统数据速率优化中，以克服BS在和数据速率提高的有利情况下的支配效应。此外，我们推导出了单个集群的 MIMO-NOMA 资源分配的封闭形式解决方案，它可以用于任何规模的集群。此外，我们提出了一种新的两侧联合匹配方法来联合优化 MIMO-NOMA 聚类和 BS 选择，这能够平衡 MU 的个体利益和整体网络性能之间的权衡。建议的方法是核心稳定的。Pauta 准则用于系统性能评估，以提供对双赢解决方案的判断。在模拟中，广泛的比较为我们提出的 MIMO-NOMA 聚类策略、相对公平性和提出的两侧联合匹配方法提供了深刻的理解。Pauta 准则用于系统性能评估，以提供对双赢解决方案的判断。在模拟中，广泛的比较为我们提出的 MIMO-NOMA 聚类策略、相对公平性和提出的两侧联合匹配方法提供了深刻的理解。Pauta 准则用于系统性能评估，以提供对双赢解决方案的判断。在模拟中，广泛的比较为我们提出的 MIMO-NOMA 聚类策略、相对公平性和提出的两侧联合匹配方法提供了深刻的理解。