蜂窝D2D异构网络中的干扰对齐技术研究

Device-to-device(D2D) communications in cellular networks has attracted much attention by virtue of advantages it offers: cell throughput enhancement, user equipment power savings, and instantaneous data rate increase. Interference management of D2D communications is usually addressed by power and resource allocation schemes, however, they are efficient to enhance the spectrum efficient. Interference alignment (IA) has been recognized as a promising technique that achieves the capacity of the multiple-input multiple-output (MIMO) interference channel for high signal to noise. This project proposes using IA techniques in a cellular/D2D heterogeneous network to manage the interference. In this research, the degrees of freedom (DoF) of cellular/D2D heterogeneous networks with centralized control and distributed control is studied by applying asymptotic interference alignment scheme. The feasibilities of linear interference alignment are investigated for cellular/D2D heterogeneous networks under different interference constraints, and corresponding interference alignment solutions are designed based on linear IA schemes. For further improving the performance of interference management for cellular/D2D heterogeneous networks, a joint user scheduling and transmission strategy selection scheme is proposed by applying stochastic networks optimization. We study the performance loss analysis of interference alignment for cellular/D2D heterogeneous networks under limited feedback at first. To enhance the performance of limited feedback, we design a dynamic feedback-bits allocation approach for each receiver device, and study feedback-topology designs based on the network structure of cellular/D2D heterogeneous networks. We also proposed a rotation hierarchical codebook to reduce the storage complexity and computation complexity of CSI quantization.

终端到终端(Device-to-Device, D2D)直通通信技术能够提高蜂窝网络吞吐量，降低用户设备功耗，提高数据的瞬时传输速率，因此得到了学术界的广泛关注。传统的蜂窝D2D通信通过功率和资源分配方式进行干扰管理，对频谱效率的提高有限。干扰对齐(Interference Alignment, IA)是一种高效的干扰管理技术，在高信噪比条件下能够使MIMO干扰信道系统达到最优性能。本项目提出基于IA技术对蜂窝D2D异构网络进行干扰管理。基于渐进IA技术进行蜂窝D2D异构网络自由度分析。研究不同干扰约束条件下蜂窝D2D异构网络线性干扰对齐的可行性条件，并提出相应的线性IA解决方案。为进一步提高蜂窝D2D异构网络干扰管理的性能，基于随机网络优化理论提出一种联合用户调度和传输策略选择机制。最后，对基于有限反馈IA进行研究。包括有限反馈对IA性能的影响，动态反馈比特分配机制和低复杂度量化码本。

用户间的干扰是蜂窝D2D异构网络面临的主要问题，其严重影响了系统的信道容量。本项目基于干扰对齐技术进行蜂窝D2D异构网络的干扰管理研究，并从线性干扰对齐方案设计、干扰受限条件下干扰对齐设计及信道状态有限反馈等三个反面进行了研究，具有重要的理论意义和广阔的应用前景。.针对蜂窝D2D异构网络结构中蜂窝通信与D2D用户间的相互干扰问题，提出了一种基于分组的干扰对齐算法，将蜂窝通信基站对D2D通信用户的干扰对齐到相同的信号空间，并根据蜂窝通信和D2D通信对传输速率的不同需求，设计相应的预编码矩阵和接收滤波矩阵，消除用户间干扰对通信的影响。仿真实验表明，在有限的天线配置条件下，本项目提出的干扰对齐算法，能够满足蜂窝D2D异构网络中对系统传输容量的不同需求，具有良好的应用价值。.在非均匀路径损耗条件下，针对由信道状态信息量化产生的干扰泄露问题，研究分析了其对系统容量的影响，得到了信道容量与量化比特数的函数关系；提出了一种自适应反馈比特分配算法，以在总反馈信息量相同的条件下，有效提高系统的容量。并且，针对MIMO干扰信道多子数据流传输情况下的信道状态信息反馈问题，分析了由于信道状态量化而导致的系统容量损失，得到了系统容量与量化比特数的函数关系及提出一种动态比特分配算法，仿真验证了理论分析结果的有效性和提出算法的性能。.针对信道状态信息有限反馈导致的干扰泄露问题，提出利用多用户MIMO系统中路径损耗的非均匀特性，通过联合动态反馈和功率分配降低泄露干扰功率，从而有效提高系统的和速率。首先从理论上分析得到了在动态反馈和功率分配条件下系统和速率的解析表达式；以系统容量提升为目标，对反馈比特分配和功率分配优化问题进行建模；并提出根据信道的准静态特性对该问题进行低复杂度求解，得到功率和反馈比特分配方案。仿真结果表明，与单独考虑信道状态动态反馈条件下的干扰对齐技术相比，本项目提出的方法能够有效提高系统和速率，具有较高的理论和应用价值。