MIMO检测与OFDM预处理算法研究

发布时间：2018-01-21 17:25

本文关键词： MIMO V-BLAST OFDM 迭代检测预处理　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】：MIMO-OFDM是新一代移动通信系统的物理层的核心关键技术。MIMO系统能够显著提升系统的频谱效率和功率效率,同时不额外占用频谱资源,OFDM系统相对于单载波系统而言能够有效对抗无线通信系统中的多径所造成的频率选择性衰落效应。本文在MIMO系统中所研究的主要方向是V-BLAST系统下的信号检测技术,对于V-BLAST系统下的信号检测技术研究,本文主要介绍了常用的检测算法,包括线性检测算法(包括ZF检测和MMSE检测),ML检测,次最优低复杂度迭代检测算法(包括PDA检测和MMSE-ISDIC检测),线性检测算法复杂度低,但性能太差,ML检测复杂度过高,常用于与其它检测算法做比较。本文重点分析低复杂度迭代检测算法,介绍了PDA检测算法的基本原理,并给出了PDA检测算法中噪声和干扰的真实PDF。对于MMSE-ISDIC检测算法,本文给出了详细推导,并且给出了MMSE-ISDIC检测算法的一种改进,这种改进是基于传统的MMSE-ISDIC检测算法所使用的噪声和干扰的方差并不精确,从而提出了一种更精确的描述,这种改进可以在一定程度上提升原检测算法的性能。OFDM技术能够有效的对抗无线通信系统中的频率选择性衰落,但当无线信道为时变时,就会产生多普勒频移,同时产生时间选择性衰落,多普勒频移会破坏OFDM子载波间的正交性,从而产生子载波间干扰。本文在OFDM系统中所研究的主要方向为OFDM在频率选择性时间选择性(即双选择性)衰落信道下的信号检测算法,包括传统检测算法和低复杂度检测算法。本文分析了双选衰落信道对OFDM符号的影响,分析了子载波间干扰的产生机制,并给出了OFDM在双选衰落信道下的系统模型。同时,介绍了几种常用的检测算法,包括线性检测算法,以及根据双选衰落信道下的OFDM系统信道矩阵的特点而直接忽略信道矩阵带外能量的低复杂度检测算法FIR-MMSE-SIC检测算法。对于OFDM低复杂度检测算法,常用的方法是直接忽略带外能量,如FIR-MMSE-SIC检测算法,这会造成一定的性能损失。对于低复杂度检测算法,当信道为莱斯信道时,本文介绍了一种发送预处理,通过发送预处理,可以使信道矩阵的能量更向对角线集中,也就是减小ICI,从而提高低复杂度检测算法的性能。同时,本文提出了一种发送接收联合预处理的方法,通过本文所提出的发送接收联合预处理,不仅可以使信道矩阵的能量向对角线方向集中,更能保证信道矩阵大部分都集中在一定大小的带状内,从而在忽略信道矩阵带外能量时不会产生太大的性能损失。经过发送接收联合预处理后,在接收端采用低复杂度检测算法进行信号检测时,会比只有发送预处理,只有接收预处理或没有预处理的OFDM系统取得更好的性能。
[Abstract]:MIMO-OFDM is the key technology of the physical layer of the new generation mobile communication system. The MIMO-OFDM system can significantly improve the spectrum efficiency and power efficiency of the system, and it does not occupy additional spectrum resources. Compared with single carrier system, OFDM system can effectively resist the frequency selective fading effect caused by multipath in wireless communication system. The main research direction of this paper is V-BLA in MIMO system. St system signal detection technology. For the research of signal detection technology in V-BLAST system, this paper mainly introduces the commonly used detection algorithms, including linear detection algorithm (including ZF detection and MMSE detection / ML detection). The suboptimal low complexity iterative detection algorithms (including PDA detection and MMSE-ISDIC detection) have low complexity but poor performance. This paper focuses on the analysis of low complexity iterative detection algorithm and introduces the basic principle of PDA detection algorithm. The real PDFs of noise and interference in the PDA detection algorithm are given. For the MMSE-ISDIC detection algorithm, the detailed derivation is given in this paper. An improvement of the MMSE-ISDIC detection algorithm is also given, which is based on the inaccuracy of the variance of noise and interference used in the traditional MMSE-ISDIC detection algorithm. Thus, a more accurate description is proposed, which can improve the performance of the original detection algorithm. OFDM can effectively combat the frequency selective fading in wireless communication systems. However, when the wireless channel is time-varying, it will produce Doppler shift and time-selective fading, which will destroy the orthogonality between OFDM subcarriers. The main research direction of this paper is the signal detection algorithm of OFDM in frequency-selective time-selective (i.e. double-selective) fading channels. This paper analyzes the influence of double-selected fading channels on OFDM symbols, and analyzes the generation mechanism of inter-carrier interference (ISI). The system model of OFDM in dual fading channel is given. At the same time, several common detection algorithms, including linear detection algorithm, are introduced. According to the characteristics of the channel matrix of OFDM system in double fading channel, the low complexity detection algorithm FIR-MMSE-SIC algorithm which ignores the energy of the channel matrix out of band is directly ignored. For OFDM, the detection algorithm of low complexity is proposed. Low complexity detection algorithm. The common method is to directly ignore the out-of-band energy, such as FIR-MMSE-SIC detection algorithm, which will cause a certain performance loss. For low-complexity detection algorithm, when the channel is a Rice channel. This paper introduces a transmission preprocessing, which can make the energy of the channel matrix more diagonal, that is, to reduce the ICI, so as to improve the performance of the low complexity detection algorithm. In this paper, a method of joint preprocessing of transmission and reception is proposed, which can not only concentrate the energy of the channel matrix in diagonal direction. It can also ensure that most of the channel matrix is concentrated in a certain size band, so that the channel matrix energy out of the band will not produce too much performance loss. After transmission and reception joint preprocessing. When the receiver uses a low complexity detection algorithm for signal detection, it will achieve better performance than the OFDM system with only sending preprocessing, receiving preprocessing or no preprocessing.
【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】：TN919.3

【相似文献】