LTE-A系统中下行用户调度算法研究

发布时间：2018-03-16 23:22

本文选题：协作传输技术　切入点：多用户　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：协作多点传输技术CoMP(Coordinated Multiple Points Transmission/Reception)作为LTE-A(Long Term Evolution-Advance)关键技术之一,近年来成为了研究热点。LTE-A系统使用多小区多用户协作传输技术,可以有效对抗小区间干扰,明显改善小区边缘用户的服务质量,最终提高系统整体性能。但是系统中边缘用户数量很多,而一个时频资源块可以同时服务的用户数量有限,为了提高系统性能,必须利用调度算法选出一组合适的用户集合进行服务,同时考虑到可实现性,调度算法复杂度不宜过高。本文简要分析了MIMO(Multi-input Multi-output)信道容量问题,研究了用于消除用户间干扰的预编码技术。对于块对角化的线性预编码,从数学角度分析了两种简单改进方法。改进的块对角化预编码减少了矩阵分解的计算量,但是系统的性能不变。本文重点研究了下行多用户调度算法,从两方面分析了几种复杂度比较低的算法并对其进行了改进。改进算法依然应用了贪心思想,每次只选择一个用户进入已选用户集,逐步缩小待选用户集,直到系统和容量开始下降或是已选用户数达到上限。但是相比基于容量的调度,改进算法更换了调度的准则,克服了基于容量调度过程由于过多奇异值的分解引起的复杂度过高问题。本文更换的调度准则从以下两方面考虑,首先假设接收端用户之间可以协作,将单用户的容量作为多用户协作时容量界限,每次只选择可以达到更高界限的用户,其次针对块对角化预编码对信道相关性敏感的问题,每次只选择和已选用户集之间信道相关性最小的用户。本文提出的改进方法减少了矩阵迭代分解操作。仿真结果表明,改进算法兼顾系统性能的同时有效降低复杂度。本文将两种经典贪心算法运用到基于用户和天线的联合调度过程中。联合调度中单个用户不再使用所有天线接收信号,可调度的用户数量增加,系统的整体性能会有所提高。联合调度中随着天线数量增多,天线之间的相关性逐渐成为影响系统容量的关键因素,本文接下来分析的两种算法都是从天线之间的相关性出发筛选用户和用户天线,首先分析了用天线夹角方法衡量待选天线和已选天线集之间相关性的方法,其次是分析了用弦距离衡量待选天线和已选天线集之间的相关性的方法,仿真分析表明,这两种算法相比基于容量的思想会有部分损失,但是低复杂度优势明显。
[Abstract]:As one of the key technologies of LTE-A(Long Term Evolution-Advancement, cooperative multipoint transmission (CoMP(Coordinated Multiple Points transmission / recovery) has become a research hotspot in recent years. LTE-A system uses multi-cell multi-user cooperative transmission technology, which can effectively resist inter-cell interference. Obviously improve the quality of service of the edge users of the cell, and ultimately improve the overall performance of the system. But there are a large number of edge users in the system, and a time-frequency resource block can serve a limited number of users at the same time, in order to improve the performance of the system, The scheduling algorithm must be used to select a suitable set of users for service, and considering the realizability, the complexity of the scheduling algorithm should not be too high. This paper briefly analyzes the channel capacity of MIMO(Multi-input Multi-Output. In this paper, the precoding technique used to eliminate interference between users is studied. For the linear precoding of block diagonalization, two simple improved methods are analyzed from the point of view of mathematics. The improved block diagonalization precoding reduces the computation of matrix decomposition. However, the performance of the system remains unchanged. This paper focuses on the downlink multi-user scheduling algorithm, analyzes several low-complexity algorithms from two aspects and improves them. Only one user is selected to enter the selected user set at a time, gradually reducing the user set until the system and capacity begin to decline or the number of selected users reaches the upper limit. However, compared with the capacity based scheduling, the improved algorithm changes the scheduling criteria. This paper overcomes the problem of high complexity caused by the decomposition of too many singular values in the capacity-based scheduling process. The proposed scheduling criteria are considered from the following two aspects: firstly, it is assumed that the users on the receiving end can cooperate with each other. The capacity of single user is regarded as the capacity limit of multi-user collaboration. Only those users who can reach a higher limit are selected at a time. Secondly, the block diagonalization precoding is sensitive to channel correlation. Only those users with the least channel correlation between the selected user set and the selected user set are selected each time. The improved method proposed in this paper reduces the matrix iterative decomposition operation. The simulation results show that, In this paper, two classical greedy algorithms are applied to the joint scheduling process based on users and antennas. In joint scheduling, a single user no longer uses all antennas to receive signals. With the increase of the number of antennas in joint scheduling, the correlation between antennas becomes the key factor that affects the capacity of the system. The two algorithms analyzed in this paper are based on the correlation between the antennas to screen the user and the user antenna. Firstly, the method of measuring the correlation between the selected antenna and the selected antenna set using the antenna angle method is analyzed. Secondly, the method of measuring the correlation between the selected antenna and the selected antenna set by chord distance is analyzed. The simulation results show that the two algorithms have some losses compared with the capacity-based idea, but the advantages of low complexity are obvious.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5

【相似文献】