多天线系统下行干扰抑制消除的研究

发布时间：2018-01-05 00:04

本文关键词：多天线系统下行干扰抑制消除的研究　出处：《电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：智能手机的普及,大大的刺激了人们对于更高通信速率、更好通信质量的迫切需求,同时也促进了无线通信的迅猛发展。小区的细分化、天线数目的增多以及异构网络技术的应用导致了小区之间的相互覆盖越来越多,这些下行多用户干扰严重地制约着无线通信的通信质量和系统性能。传统的干扰抑制方案中常用的方法是预编码技术。由于在实际的通信过程中,发射机无法掌握下行信道状态信息,需要用户在上行信息中添加与信道信息相关的导频信号来进行信道或预编码矩阵的估计。因此预编码方法不仅受到信道估计误差的影响,还受到用户反馈误差的影响。鉴于这个问题,一些学者提出了接收机干扰处理方案的思想。此外,在预编码方案中,非线性预编码方案比线性预编码方案的系统性能要好,但是由于其复杂的计算方式导致其很少应用在实际的通信过程中。然而在点对多点的通信系统中,采用一个复杂度较高的基站比采用多个复杂度较高的接收终端更加可取。因此,一些学者提出了较低复杂度的非线性预编码方案研究方向。针对这两个问题,本文主要研究了两个干扰抑制消除方案。一个是接收机干扰消除方案,主要采用联合迭代译码技术;另外一个是发射机干扰抑制方案,主要采用扩展星座脏纸编码技术。本文首先具体地介绍了联合迭代译码的算法流程,其主要思想是:将除自身信号外的干扰信号视为加性高斯白噪声,采用特殊的译码结构将自身信号和干扰信号同时译出。由于联合迭代译码的发射机不需要任何与信道状态信息相关的处理,因此其不受信道反馈误差的影响,并且其干扰消除性能比大部分预编码方案更好。此外,本文还推导了信道估计误差方差与联合迭代译码的信号干扰噪声比值关系的闭合公式,该公式可作为实际通信过程中参数调整的参考。然后,本文详细地介绍了扩展星座脏纸编码技术的处理过程,其主要思想是:将原始发送信号在二维平面进行扩展,然后在扩展星座子集中寻找与已知干扰最接近的星座点,最后将这两点的差值作为脏纸编码符号。这样处理的方式,不仅可以降低编码后信号的峰值功率和平均功率,还能避免一些对称取模运算带来的性能损失。此外,本文还介绍了扩展星座脏纸编码技术的用户选择方法,与未采用用户选择的方案相比,采用用户选择方案在系统误码率性能上有较大的提升。
[Abstract]:The popularity of smart phones has greatly stimulated the urgent need for higher communication rate and better communication quality, but also promoted the rapid development of wireless communications. The increasing number of antennas and the application of heterogeneous network technology lead to more and more mutual coverage between cells. These downlink multiuser interference seriously restrict the communication quality and system performance of wireless communication. The common method of traditional interference suppression scheme is precoding technology. The transmitter is unable to master downlink channel state information. Users are required to add pilot signals related to channel information to estimate the channel or precoding matrix, so the precoding method is not only affected by channel estimation errors. It is also affected by user feedback error. In view of this problem, some scholars put forward the idea of receiver interference processing scheme. In addition, in the precoding scheme. The system performance of nonlinear precoding scheme is better than that of linear precoding scheme, but because of its complex calculation method, it is rarely used in the actual communication process, however, in the point-to-multipoint communication system. It is preferable to adopt a base station with a high complexity than a receiver terminal with a higher complexity. Therefore, some scholars have proposed a low complexity nonlinear precoding scheme. In order to solve these two problems, some scholars have proposed the research direction of nonlinear precoding scheme with lower complexity. In this paper, two interference cancellation schemes are studied. One is the receiver interference cancellation scheme, which mainly adopts the joint iterative decoding technique. The other is the transmitter interference suppression scheme, which mainly adopts the extended constellation dirty paper coding technology. Firstly, this paper introduces the algorithm flow of joint iterative decoding in detail. Its main idea is: besides its own signal interference signal is regarded as additive Gao Si white noise. Because the transmitter of joint iterative decoding does not need any processing related to the channel state information, it is not affected by the channel feedback error. Moreover, the interference cancellation performance is better than that of most precoding schemes. In addition, the closed formula of the relationship between channel estimation error variance and signal interference noise ratio of joint iterative decoding is derived. The formula can be used as a reference for parameter adjustment in the process of practical communication. Then, this paper introduces the processing process of the extended constellation dirty paper coding technology in detail. The main idea is to extend the original transmission signal in the two-dimensional plane, and then to find the constellation closest to the known interference in the extended constellation subset. Finally, the difference between these two points is used as the dirty paper coding symbol. In this way, not only the peak power and average power of the coded signal can be reduced. In addition, this paper also introduces the user selection method of extended constellation dirty paper coding technology, compared with the scheme without user selection. The BER performance of the system is greatly improved by using the user selection scheme.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN820;TN972

【相似文献】