Massive MIMO环境下分布式信源的波达方向估计算法研究

发布时间：2018-02-20 17:12

本文关键词： 阵列信号处理波达方向估计 Massive MIMO系统相干分布式信源　出处：《内蒙古大学》2017年硕士论文　论文类型：学位论文

【摘要】：波达方向(Direction of Arrival,DOA)估计是用来获得信号到达角度的一门学科,其在工程实践中也被受关注。在无线移动通信系统中,DOA估计起到了对移动用户跟踪定位的关键作用,因此对波达方向估计算法的研究在理论与实践中都具有重要的意义。传统的DOA估计算法假设信号模型为点信源模型,但是随着人们的不断研究,发现实践中传输的信号受多径效应的影响,使信号的能量被扩散,不能按照一点入射。因此,本文为了能与实际相结合将采取分布式信源模型来模拟上述情景。此外,在实际应用中,算法处理的实时性也至关重要,因此,本课题在相干分布式信源模型下,提出了一种可以在实数域处理的快速DOA估计新方法(M-U-ESPRIT算法),该算法将酉ESPRIT算法与可以快速降低维度的多级维纳滤波器相结合,省略了对数据协方差矩阵求特征值的过程,在保证了估计性能的前提下能够有效地减少算法在计算过程中的运算量,达到快速估计信号到达角的目的。在无线移动通信领域,Massive MIMO系统由于具有极高的频谱利用率和较高的接收功率等优点,在近五年内得到了迅猛发展,已经成为目前的前沿研究方向。本文将第三章研究的基于相干分布式信源模型的DOA估计算法应用在Massive MIMO系统中。在基于均匀矩形阵的Massive MIMO系统中,本文推导了服从高斯分布的二维相干分布式信源方向向量的近似表达式。在此基础上,本文将酉ESPRIT算法和新方法M-U-ESPRIT算法分别应用在Massive MIMO系统中,得到了二维CDU-ESPRIT算法和二维CDM-ESPRIT算法,针对这两种新算法本文在各种不同的实验环境下做了大量的Monte Carlo仿真。实验结果说明这两种算法能有效的估计出待估计仰角和方位角,并且在不同的实验条件下都具有良好的估计性能。
[Abstract]:Direction of arrival (DOA) estimation is a discipline used to obtain the angle of arrival of signals, which is also concerned in engineering practice. In wireless mobile communication systems, DOA estimation plays a key role in tracking and locating mobile users. Therefore, the research on DOA estimation is of great significance in both theory and practice. The traditional DOA estimation algorithm assumes that the signal model is a point source model, but with the continuous research of people, It is found that the signal transmitted in practice is affected by multipath effect, so that the energy of the signal is diffused and can not be incident at one point. Therefore, in order to be able to combine with practice, the distributed source model will be adopted to simulate the above situation. In the practical application, the real-time processing of the algorithm is also very important. Therefore, under the coherent distributed source model, A new fast DOA estimation method, M-U-ESPRIT algorithm, is proposed, which can be processed in real number domain. This algorithm combines unitary ESPRIT algorithm with multistage Wiener filter which can quickly reduce the dimension, and omits the process of finding the eigenvalue of the data covariance matrix. On the premise of ensuring the performance of the estimation, the computational complexity of the algorithm can be reduced effectively. In the field of wireless mobile communication, passive MIMO system has been developed rapidly in the past five years because of its high spectrum efficiency and high receiving power. In chapter 3, the DOA estimation algorithm based on coherent distributed source model is applied to Massive MIMO system. In this paper, the approximate expression of directional vector of two-dimensional coherent distributed information source with Gao Si distribution is derived. On this basis, the unitary ESPRIT algorithm and the new method M-U-ESPRIT algorithm are applied to Massive MIMO system, respectively. The two-dimensional CDU-ESPRIT algorithm and two-dimensional CDM-ESPRIT algorithm are obtained, and a large number of Monte Carlo simulations are done in different experimental environments for these two new algorithms. The experimental results show that these two algorithms can effectively estimate the elevation and azimuth angles to be estimated. And it has good estimation performance under different experimental conditions.
【学位授予单位】：内蒙古大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5;TN911.7

【相似文献】