异构云小蜂窝网络中基于协作预编码的干扰协调技术研究

The heterogeneous cloud small cell network(HCSCNet), which combines cloud computing and a heterogeneous small cell network, will likely play an important role in 5G mobile communication network. However, with massive deployment of small cells, co-channel interference is an important problem in an HCSCNet, especially for cell edge users. Interference coordination approach mainly include coordinated multi-point transmission/reception(CoMP) and interference alignment(IA). The concrete realization of the CoMP and IA mainly depends on the precoding technique.. This project is to the mobile user classification and small-cells clustering scheduling and precoding design in HCSCNet to deal with co-channel interference problem. Mainly combined with the spectrum efficiency and QoS requirements of users, the present project designs real-time user classification and small-cells clustering scheduling and the linear or near linear precoding design algorithm. The concrete content includes: firstly, design the user classification and small-cells clustering scheduling algorithm with real-time according to the limited user information analysis; secondly, based on CoMP and IA technology, study the linear or near linear precoding design algorithm for a static small-cells clustering under the perfect channel state information; thirdly, study the linear or near linear precoding design algorithm for semi-dynamic clustering and full-dynamic clustering, respectively, under the perfect channel state information; fourthly, study the linear precoding design algorithm with robustness under the imperfect channel state information; and finally, verify the proposed theories and methods through system simulations studies.

云计算融入异构小蜂窝网络中所形成的异构云小蜂窝网络（HCSCNet）是最有希望成为5G通信网络架构。然而，小蜂窝的大规模部署势必会造成严重的同频信道干扰，尤其是对蜂窝边缘用户而言。干扰协调的处理方式主要有协作多点收发和干扰对齐，具体实现主要是依靠预编码技术。.本项目以HCSCNet中的用户分类和蜂窝成簇调度及相应的预编码设计为研究对象以处理其中的同频信道干扰问题，主要是结合频谱效率与用户的QoS要求，设计实时的用户分类与蜂窝成簇调度算法、以及相应的线性的或接近线性的预编码设计算法。具体包括：1）如何根据用户信息设计实时的用户分类和蜂窝成簇调度算法；2）基于多点收发和干扰对齐技术，研究蜂窝静态成簇的情况下线性或接近线性的预编码设计算法；3）研究蜂窝动态/半动态成簇的情况下线性或接近线性的预编码设计算法；4）研究不完备信道信息情况下具有鲁棒性的线性预编码设计算法；最后建立系统级仿真验证。

云计算融入异构小蜂窝网络中所形成的异构云小蜂窝网络（HCSCNet）是5G及B5G通信网络的首选架构。然而，小蜂窝的大规模部署势必会造成严重的同频信道干扰，尤其是对蜂窝边缘用户而言。干扰协调的处理方式主要有协作多点收发和干扰对齐，具体实现主要是依靠预编码技术、载波选择方案等。.基于此，本项目开展的研究内容包括：.1）MIMO中继广播信道中的联合信源与中继的预编码设计机制；2）基于SLNR的MU-MIMO系统预编码和用户调度机制；3）基于人工蜂群算法的基站天线方位角与下倾角自优化方法；4）融合通用滤波多载波的非正交多址接入技术；5）OFDM–IM系统中的载波激活选择策略；6）Relay 系统中基于相关信源的联合迭代译码机制。在外，受到AI在无线通信领域快速发展的启发和影响，项目组成员还在移动边缘计算中的多用户卸载算法问题方面开展了一些自由性的探索研究，并取得了一定的成果。.项目的研究成果将为异构云小蜂窝网络（HCSCNet）在现实中的应用提供技术支持，为干扰处理的相关理论和技术提供重要参考。