基于压缩感知的OFDM系统信道估计算法研究

发布时间：2018-04-03 11:37

本文选题：OFDM　切入点：信道估计　出处：《天津工业大学》2017年硕士论文

【摘要】：正交频分复用(Orthogonal Frequency Division Multiplexing,OFDM)技术已经成为无线通信技术中不可替代的部分,具有很高的应用价值。信道估计是OFDM通信系统中的关键技术,信道估计性能的好坏将直接影响到整个系统的通信质量。'压缩感知理论的提出,有效的改善了 OFDM系统稀疏信道估计的性能,减少了信道估计所需要的导频数,提高了系统的频谱利用率。本文分析了基于压缩感知的OFDM系统信道估计问题。在OFDM系统中,如果信道的稀疏度是已知的,传统的压缩感知算法,如正交匹配追踪算法(Orthogonal Matching Pursuit,OMP)、压缩采样匹配追踪算法(Compressive Sampling Matching Pursuit,CoSaMP)、子空间追踪(Subspace Pursuit,SP)在合理的参数选取下均能表现出良好的估计性能,表明了压缩感知算法在OFDM系统信道估计中优越性。然而,由于在实际系统中,信道的稀疏度通常是未知的,极大的限制了需要预知稀疏度的压缩感知算法在OFDM信道估计中的实际应用。为了更好的利用压缩感知去实现OFDM信道估计,需要研究自适应稀疏度的恢复算法。文章先介绍了传统的稀疏度自适应匹配追踪(Sparsity Adaptive Matching Pursuit,SAMP)算法信道估计,而SAMP算法虽然可以达到自适应稀疏度的效果,但由于存在欠估计和过估计的问题,给信道估计的性能带来了较为不利的影响,同时,为了追求更好的性能,需要提高算法的计算复杂度,极大的影响了通信系统的实时性。针对以上算法的不足,本文提出了一种正则化自适应稀疏度的压缩感知算法(Regularized Sparsity Adaptive Matching Pursuit,RSAMP),该算法不需要预先知道信道的稀疏度,首先通过选择相关系数向量中最大后向差分的位置来选择支撑集原子,再对已选择的原子支撑集合进行正则化,用来提高支撑集的准确性,并通过迭代直至算法收敛。在未知信道稀疏度时,算法有着良好的性能,并具有较低的计算复杂度。同时,由于OFDM系统中较高的峰均比影响了功率放大器的工作性能,如果对OFDM系统进行限幅操作来抑制高峰均比,就会使导频信号受到非线性失真的影响,从而严重影响信道估计性能。针对这一问题,本文提出了利用迭代的方法,用压缩感知对信道响应和非线性失真分别进行估计,通过对导频信号进行补偿,减小非线性失真对信道估计性能的影响,扩展了压缩感知在OFDM系统信道估计中的应用场景。
[Abstract]:Orthogonal Frequency Division multiplexing (OFDM) technology has become an irreplaceable part of wireless communication technology and has high application value.Channel estimation is a key technology in OFDM communication system. The performance of channel estimation will directly affect the communication quality of the whole system.The proposed compressed sensing theory can effectively improve the performance of sparse channel estimation in OFDM systems, reduce the number of pilots needed for channel estimation, and improve the spectral efficiency of the system.In this paper, the problem of channel estimation for OFDM systems based on compressed sensing is analyzed.In OFDM systems, if the channel sparsity is known, the traditional compression sensing algorithm,For example, orthogonal Matching pursuit algorithm, compressed Sampling Matching pursuit algorithm, subspace tracker subspace pursuit algorithm can all show good estimation performance under reasonable parameter selection, which shows the superiority of compressed sensing algorithm in channel estimation of OFDM system.However, because the channel sparsity is usually unknown in the actual system, it greatly limits the practical application of the compression sensing algorithm which needs to predict the sparse degree in OFDM channel estimation.In order to make better use of compressed sensing to realize OFDM channel estimation, it is necessary to study an adaptive sparse recovery algorithm.This paper first introduces the channel estimation of the traditional sparse adaptive matching tracking Adaptive Matching pursuit algorithm. Although the SAMP algorithm can achieve the effect of adaptive sparsity, it has the problem of underestimation and overestimation.At the same time, in order to achieve better performance, it is necessary to improve the computational complexity of the algorithm, which greatly affects the real-time performance of the communication system.To overcome the shortcomings of the above algorithms, a regularized Sparsity Adaptive Matching pursuit algorithm is proposed in this paper. The algorithm does not need to know the sparse degree of the channel in advance.Firstly, the support set atom is selected by selecting the position of the largest backward difference in the correlation coefficient vector, and then the selected atomic support set is regularized to improve the accuracy of the support set, and then iterate until the algorithm converges.When the channel sparsity is unknown, the algorithm has good performance and low computational complexity.At the same time, because the high peak-to-average ratio (PAPR) in the OFDM system affects the performance of the power amplifier, if the OFDM system is limited to suppress the PAPR, the pilot signal will be affected by nonlinear distortion.Thus the channel estimation performance is seriously affected.To solve this problem, an iterative method is proposed to estimate channel response and nonlinear distortion separately by compression sensing, and to reduce the influence of nonlinear distortion on channel estimation performance by compensating pilot signals.The application of compressed sensing in OFDM channel estimation is extended.
【学位授予单位】：天津工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.53

【参考文献】