大规模MIMO预编码算法及其应用研究

发布时间：2018-05-05 19:30

本文选题：大规模MIMO + 预编码　；参考：《南京邮电大学》2017年硕士论文

【摘要】：随着无线移动通信的迅速发展,人们对无线移动通信中的速率、延时和吞吐量等方面的要求也不断提高。但传统的MIMO技术已不能满足这一要求,于是大规模MIMO技术被提出,它可以提高整个系统的频谱利用率和吞吐量等。由于在大规模MIMO系统中,基站部署数百根天线,并服务更多的用户,因此用户之间的干扰会更大,解决这个问题的方法之一就是预编码技术。本文研究了大规模MIMO系统中的预编码算法设计的问题,分别针对完美信道状态信息(CSI)、非完美CSI和非理想硬件三种不同场景,通过分析特定问题对系统的影响和传统预编码算法的不足,提出一种优化的预编码算法。论文研究内容如下:针对完美CSI的情况下,我们研究了单个小区和多个小区场景中的预编码问题。对于单个小区场景,通过结合传统迫零(ZF)和块对角化(BD)预编码算法,提出了一种优化的BD预编码算法。对于多个小区场景,我们分析了小区间干扰形成的原因,然后,利用基于每根接收天线的信号与泄漏加噪声比准则来得到预编码矩阵,从而能提高小区边缘用户的性能。仿真结果表明,本文算法不但可以使得系统容量有所提升,而且也有着更低的系统误码率。针对非完美CSI的情况下,我们主要研究了在单个小区场景下,分析了非完美CSI对系统的影响,我们利用两种方法来设计预编码矩阵,包括单用户的最小均方误差和最大信干比准则。另外,本节除了研究瑞利信道模型下的预编码设计问题,还简单介绍了莱斯信道模型和Nakagami信道模型,并对这三种信道模型进行了仿真比较。仿真结果表明,本文算法能够在一定程度上降低非完美CSI对系统影响程度。针对非理想硬件的问题,我们研究了相位正交(IQ)失衡、高功率放大器(HPA)非线性和相位噪声三个问题。对于这三个问题,我们首先分析了其对系统的影响。其次,给出传统预编码算法的不足。最终提出优化预编码算法。结合仿真结果,我们可知本文算法可以在一定程度上弥补非理想硬件的不足。针对中继系统中的预编码算法的设计,首先,本文介绍中继系统的相关概念。其次,本文使用最小均方误差准则且结合本文理论部分知识得出预编码矩阵,在求解的过程中减少未知量数,简化代价函数,降低了计算复杂度。
[Abstract]:With the rapid development of wireless mobile communication, the requirements of rate, delay and throughput in wireless mobile communication are also increasing. However, the traditional MIMO technology can not meet this requirement, so large-scale MIMO technology is proposed, which can improve the spectrum efficiency and throughput of the whole system. In large scale MIMO systems, the base station deploys hundreds of antennas and serves more users, so the interference between users will be greater. One of the methods to solve this problem is pre-coding technology. In this paper, we study the design of precoding algorithm in large-scale MIMO system, which aims at three different scenarios: perfect channel state information, imperfect CSI and imperfect hardware, respectively. By analyzing the influence of specific problems on the system and the shortcomings of traditional precoding algorithms, an optimized precoding algorithm is proposed. The main contents of this paper are as follows: in the case of perfect CSI, we study the precoding problem in single cell and multiple cell scenarios. For a single cell scenario, an optimized BD precoding algorithm is proposed by combining the traditional zero-forcing ZF algorithm with block diagonalization (BDD) precoding algorithm. For multiple cell scenarios, we analyze the cause of inter-cell interference, and then obtain the precoding matrix by using the ratio criterion of signal to leakage and noise based on each receiving antenna, which can improve the performance of cell edge users. Simulation results show that the proposed algorithm not only improves the system capacity, but also has a lower bit error rate. In the case of imperfect CSI, we mainly study the effect of imperfect CSI on the system in a single cell scenario. We use two methods to design the precoding matrix. The minimum mean square error and maximum signal-to-interference ratio criterion of single user are included. In addition, this section not only studies the design of precoding under Rayleigh channel model, but also briefly introduces Rice channel model and Nakagami channel model, and makes a simulation comparison of these three channel models. Simulation results show that the proposed algorithm can reduce the influence of imperfect CSI on the system to some extent. In order to solve the problem of non-ideal hardware, we study three problems: phase quadrature IQ) imbalance, high power amplifier (HPA) nonlinearity and phase noise. For these three problems, we first analyze their impact on the system. Secondly, the deficiency of the traditional precoding algorithm is given. Finally, an optimized precoding algorithm is proposed. Combined with the simulation results, we know that the algorithm can make up for the deficiency of non-ideal hardware to some extent. For the design of precoding algorithm in relay system, firstly, this paper introduces the related concepts of relay system. Secondly, the minimum mean square error criterion is used to obtain the precoding matrix, which reduces the number of unknown variables, simplifies the cost function and reduces the computational complexity.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.3

【参考文献】