MIMO干扰网络中面向干扰和资源管理的有限字符预编码研究

Limited resources and interference are known as two main factors which limit the improvement of the overall system performance of the wireless communication networks. Precoding is frequently employed in Multi-user MIMO systems since it can pre-process signals to be transmitted according to the channel sate information (CSI), and thus mitigate the inter-user interference. Procoding combined with power control and resource allocation is also good scheme managing interference. However, those traditional optimal precoding theory, which is based on the assumption that the input signals are Gaussian distributed, may no longer be optimal any more for real communication networks where the input signals are selected from a set with discrete finite alphabet elements. With the objective to maximize the mutual information between the input and output under the finite alphabet constraints in interference networks, we will study precoding schemes which combine interference management and resource allocation to achieve the optimal transmission of information under the constraints of practical constellation inputs. To achieve this, efforts will be made from three interrelated perspective, i.e., the achievable rate region for interference networks with finite alphabets input, the optimization theory of precoding design with finite alphabet constraints and the joint transceiver design theory for the practical implementation. The objective of the project is to find and improve the transmit theory frame and techniques for the finite alphabet constrained interference networks. Through the exploration to the creative ideas of the modeling of interference network with finite alphabet constraint, the distributed precoding design and the utilizing of long-term channel statistical information, we hope to realize interference and resource management oriented precoding schemes, and make them practical for real communication networks.

有限的资源和干扰是制约无线通信网络性能提升的两大关键因素。MIMO多用户通信中的预编码技术可根据信道状态信息对待发送的信号进行预处理，并可结合功率控制以及无线资源分配进行干扰管理。但传统的基于高斯分布输入的最优预编码设计理论对于离散有限字符输入的实际通信系统而言将不再是最优。本项目将以实现MIMO干扰网络中有限字符约束的输入输出互信息最大化为目标，研究干扰管理与资源分配相结合的预编码方案，建立并完善实际有限字符约束的干扰网络传输理论框架与技术，以期实现实际星座约束下的最优化信息传输。基于此，本项目将重点研究有限字符约束下干扰网络的可实现速率区域问题、预编码优化设计理论和面向实际应用的收发机联合优化设计理论，并通过对有限字符约束的干扰网络建模、分布式预编码设计和利用信道统计信息等创新性思想的探索，实现面向干扰和资源管理的预编码设计方案。

多天线（MIMO）技术通过多个天线实现多发多收，可充分利用空间资源，在不增加频谱资源和天线发射功率的情况下，提高无线通信系统的有效性和可靠性，因此被视为未来移动通信的核心技术。在多用户MIMO网络中，有限的资源以及干扰是制约系统性能提升的关键因素。基于此，项目组面向多用户MIMO系统，围绕有限字符集约束的系统容量分析、基于有限字符集的新型空间调制多天线传输方案设计、MIMO干扰网络中干扰管理技术以及新型无线网络架构下的无线资源管理优化配置等进行了研究，代表性研究点有：MIMO干扰网络以及空间调制系统容量分析、MISO系统保密容量分析、有限字符约束的新型三维星座图设计、低反馈自适应比特映射方案、基于空间调制的无线信息与能量同传系统设计、MIMO网络和全双工中继系统干扰管理、无线视频传输网络和车联网场景的无线资源联合优化配置等。项目组共发表论文33篇，其中期刊论文16篇，包括IEEE Transactions 9篇（IEEE Transactions on Vehicular Technology，IEEE Transactions on Circuits & Systems for Video Technology，IEEE Transactions on Communications，IEEE Transactions on Wireless Communications等）以及IEEE Communications Letters等通信领域国际顶级期刊；会议论文17篇，包括IEEE ICC，IEEE GLOBECOM等通信领域国际旗舰会议。申请专利6项，其中授权1项。部分成果“无线移动通信基础传输理论与技术”获得山东省自然科学奖二等奖。项目组的研究成果为下一代无线通信系统的发展提供了理论支撑和技术参考。