With the rapid development of future wireless communication, the combination of NOMA technology and millimeter-wave(mmWave) technology has become a research hotspot. The application of NOMA in mmWave heterogeneous networks can meet the diverse needs of users in different applications and scenarios in future communications. In this paper, we propose a machine learning framework to deal with the user association, subchannel and power allocation problems in such a complex scenario. We focus on maximizing the energy efficiency (EE) of the system under the constraints of quality of service (QoS), interference limitation, and power limitation. Specifically, user association is solved through the Lagrange dual decomposition method, while semi-supervised learning and deep neural network (DNN) are used for the subchannel and power allocation, respectively. In particular, unlabeled samples are introduced to improve approximation and generalization ability for subchannel allocation. The simulation indicates that the proposed scheme can achieve higher EE with lower complexity.

中文翻译：

---

NOMA 网络中基于深度学习的无线电资源管理：用户关联、子信道和功率分配

随着未来无线通信的快速发展，NOMA技术与毫米波（mmWave）技术的结合成为研究热点。NOMA在毫米波异构网络中的应用，可以满足用户在未来通信中不同应用和场景下的多样化需求。在本文中，我们提出了一个机器学习框架来处理这种复杂场景中的用户关联、子信道和功率分配问题。我们专注于在服务质量 (QoS)、干扰限制和功率限制的约束下最大化系统的能源效率 (EE)。具体来说，用户关联是通过拉格朗日对偶分解方法解决的，而半监督学习和深度神经网络（DNN）分别用于子信道和功率分配。特别地，未标记样本被引入以提高子信道分配的近似和泛化能力。仿真表明，所提出的方案可以以较低的复杂度实现更高的 EE。