基于压缩感知贪婪算法的MIMO-OFDM系统信道估计研究

发布时间：2018-04-11 17:49

本文选题：MIMO-OFDM + 信道估计　；参考：《江西理工大学》2017年硕士论文

【摘要】：现阶段随着各领域技术的整体发展,在通讯领域中传统衡量人们通信质量的标准已不再适用,现代通信正在朝着高质量、高速率的方向发展。在这种飞速发展的大环境下,必须对传统的通信技术进行突破,在传统技术的基础上进行改进或者寻找新的满足现阶段通信的关键技术。以MIMO-OFDM系统为框架的现代通信系统,通过使用天线的多收多发技术,使该系统具有诸多优点,能够满足现阶段高质量、高速率的通信传输标准。但是在实际信号传输过程中信号会受到周围环境以及障碍物的影响,产生不同程度的衰落和时延,造成符号间的干扰,影响信号的传输质量。本文围绕MIMO-OFDM系统为框架,采用全新的压缩采样技术对稀疏多径信道进行研究。以达到获取更加精确的CSI,提升通信传输质量和速率的目的。本文主要完成的工作有:(1)对MIMO-OFDM系统模型展开研究,在讨论了其系统模型和基本原理的基础上,并研究了信道的衰落特性和导频结构。为了克服衰落特性对传输的影响,对基于导频的LS、MMSE和基于DFT的算法进行分析研究,并在MIMO-OFDM系统中对这几种算法进行仿真,分析其优缺点。(2)对多用户信道估计进行了研究分析,总结了多用户MIMO-OFDM系统中的编码和波束赋型技术,并对其信道容量进行分析。由于多用户在获取信道状态信息时存在用户间的干扰问题,系统获取的信道状态信息是不完整的,本文在MMSE算法的基础上利用信道的时延包络特性对信道状态信息进行进一步的精确。仿真表明改进后的算法在对多用户MIMO-OFDM系统的信道估计在精确度上有所提高。(3)理论分析和实测表明信道具有稀疏特性,针对信道的这种特性,对基于CS理论的贪婪算法进行信道估计研究。重点讨论了基于压缩感知的几种贪婪算法:OMP、ROMP及SP算法,并给出了实验仿真比较。同时根据压缩感知算法的稀疏度和复杂度问题,本文在对CoSaMP算法进行研究的基础上,根据原子弱选择标准对CoSaMP算法进行改进,提出了基于阈值改进的CoSaMP算法,仿真表明改进后的算法提高了算法的速率。
[Abstract]:At present, with the overall development of various fields of technology, the traditional standard of measuring people's communication quality is no longer applicable in the field of communication, and modern communication is developing towards the direction of high quality and high speed.In this environment of rapid development, we must break through the traditional communication technology, improve on the basis of the traditional technology or find new key technologies to meet the current stage of communication.The modern communication system based on MIMO-OFDM system has many advantages by using the technique of multiple antennas, which can meet the communication standards of high quality and high speed at the present stage.However, in the actual signal transmission process, the signal will be affected by the surrounding environment and obstacles, resulting in varying degrees of fading and delay, causing inter-symbol interference, and affecting the quality of signal transmission.In this paper, a novel compressed sampling technique is used to study sparse multipath channel around MIMO-OFDM system.In order to achieve more accurate CSI, improve the quality and rate of communication transmission.The main work of this paper is to study the MIMO-OFDM system model. Based on the discussion of the system model and the basic principle, the fading characteristics and pilot structure of the channel are studied.In order to overcome the influence of fading characteristics on transmission, the algorithms based on pilot frequency and DFT are analyzed and simulated in MIMO-OFDM system. The advantages and disadvantages of these algorithms are analyzed.The coding and beamforming techniques in multiuser MIMO-OFDM systems are summarized, and the channel capacity is analyzed.The channel state information obtained by the system is incomplete because of the interference between the users when the multi-user acquires the channel state information.Based on the MMSE algorithm, the channel state information is further accurate by using the time-delay envelope of the channel.Simulation results show that the improved algorithm improves the accuracy of channel estimation for multi-user MIMO-OFDM systems. The theoretical analysis and experimental results show that the channel has sparse characteristics.The greedy algorithm based on CS theory is studied for channel estimation.This paper mainly discusses several greedy algorithms based on compression perception: OMP ROMP and SP algorithms, and gives the experimental simulation comparison.At the same time, according to the problem of sparse degree and complexity of compressed sensing algorithm, based on the research of CoSaMP algorithm, this paper improves the CoSaMP algorithm according to the atomic weak selection criterion, and proposes an improved CoSaMP algorithm based on threshold.Simulation results show that the improved algorithm improves the speed of the algorithm.
【学位授予单位】：江西理工大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN919.3

【参考文献】