基于压缩感知的星地混合OFDM系统信道估计方法

发布时间：2018-06-18 00:36

本文选题：星地混合信道 + 压缩感知技术　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：星地混合OFDM通信系统具有广覆盖、通信距离远、地理限制小、组网灵活等优点。无线信号在星地混合信道的传输过程中,会经历路径损耗、阴影衰落、多径效应、多普勒频移等影响,严重影响信号的传输质量。为了保证接收机部分的信道均衡与相干解调,需要通过关键的信道估计技术获得信道状态信息,以保证信号的传输质量。随着压缩感知理论的提出,对稀疏信号的采样实现了以远低于奈奎斯特采样速率的巨大飞跃。考虑到实际的OFDM多径信道在时域上具有的稀疏特性,将压缩感知应用于星地混合OFDM信道的估计方法中,与传统信道估计方法相比,能显著降低导频开销,提升了频谱利用率,并获得良好的信道估计性能。本文充分利用了星地混合OFDM系统中信道所具有的稀疏性,即只有几个少数个径具有较大增益的特点,利用压缩感知技术完成信道估计,为实现星地子系统间透明的空中接口提供技术支撑。文中主要研究了基于压缩感知的信道估计方法,对导频的插入方式和信道状态信息恢复算法这两方面着手,即对“探测”和“重构”过程分别进行优化,前者影响了信道的估计效果,而后者影响了CSI的准确程度和重构速率。本文的研究内容如下:第一,对OFDM系统时域的稀疏信道进行建模,并将压缩感知的理论与信道估计问题相联系。与传统信道估计方法的估计性能相对比,选用随机分布的导频子载波探测星地混合信道,在重构部分采用正交匹配追踪算法。通过对两种典型的星地混合信道模型进行估计,验证了压缩感知信道估计方法的可行性,以理论及数据两种形式,为导频数目的选取和后续重构算法的设计提供支持。第二,对导频子载波索引的选取方法进行优化,从信道探测方面提升估计性能。先提出导频位置选取决定恢复矩阵的结论,再基于最小化列相关准则和改进准则对稀疏信道进行估计。通过对星地混合信道的仿真实验,验证了这两种准则的性能优势,得到了不同信道场景下的优质导频图样,为高效信道重构提供有力支撑。第三,对基于压缩感知的信道估计重构算法进行改进,从信道重构方面提升估计性能。针对OMP等重构算法不能在稀疏度未知的场景进行信道估计的缺陷,对稀疏自适应匹配追踪(SAMP)算法进行研究,通过对该算法的优缺点分析,针对迭代停止条件、步长选取方式和索引更新门限提出了改进的AS-SAMP算法和AS-RSAMP算法。这两种算法既保证了SAMP算法不被稀疏度未知所局限的优势,又优化了SAMP算法的性能。通过仿真实验,验证了算法在对星地混合信道进行估计时在运算速度和重构精度上的优势。
[Abstract]:The satellite-ground hybrid OFDM communication system has the advantages of wide coverage, long communication distance, small geographical limitation, flexible networking and so on. During the transmission of wireless signal in the space-ground mixed channel, the transmission quality will be seriously affected by path loss, shadow fading, multipath effect and Doppler frequency shift. In order to ensure the channel equalization and coherent demodulation of the receiver, it is necessary to obtain the channel state information through the key channel estimation techniques to ensure the transmission quality of the signal. With the development of compression sensing theory, the sampling of sparse signals has achieved a great leap forward, which is far lower than the sampling rate of Nyquist. Considering the sparse characteristic of OFDM multipath channel in time domain, compression sensing is applied to the estimation of space-ground hybrid OFDM channel. Compared with the traditional channel estimation method, it can significantly reduce the pilot overhead and improve the spectrum efficiency. Good channel estimation performance is obtained. In this paper, we make full use of the sparsity of the channel in the space-ground hybrid OFDM system, that is, only a few paths have a large gain, and use the compression sensing technique to complete the channel estimation. It provides technical support for the realization of transparent air interface between satellite and earth subsystems. In this paper, the channel estimation method based on compressed sensing is studied, and the two aspects of pilot insertion and channel state information recovery algorithm are studied, that is, the "detection" and "reconstruction" processes are optimized, respectively. The former affects the channel estimation, while the latter affects the accuracy and reconstruction rate of CSI. The main contents of this paper are as follows: firstly, the sparse channel in time domain of OFDM system is modeled, and the theory of compressed sensing is associated with the channel estimation problem. Compared with the traditional channel estimation method, the randomly distributed pilot subcarrier is used to detect the space-ground mixed channel, and the orthogonal matching tracking algorithm is used in the reconstruction part. The feasibility of compressed perceptual channel estimation is verified by estimating two typical space-ground mixed channel models, which can support the selection of pilot number and the design of subsequent reconstruction algorithm in both theoretical and data forms. Secondly, the selection method of pilot subcarrier index is optimized to improve the performance of channel detection. First, the decision recovery matrix of pilot location selection is proposed, and then the sparse channel is estimated based on the minimization column correlation criterion and the improved criterion. The performance advantages of these two criteria are verified by simulation experiments on the space-ground mixed channel, and the high quality pilot patterns under different channel scenarios are obtained, which provide a strong support for efficient channel reconstruction. Thirdly, the channel estimation reconstruction algorithm based on compressed sensing is improved to improve the performance of channel reconstruction. In view of the defect that OMP and other reconstruction algorithms can not estimate the channel in scenes with unknown sparsity, the sparse adaptive matching tracking (SAMP) algorithm is studied. Through the analysis of the advantages and disadvantages of the algorithm, the iterative stopping conditions are analyzed. An improved AS-SAMP algorithm and an AS-RSAMP algorithm are proposed for step size selection and index update threshold. These two algorithms not only guarantee the advantages of SAMP algorithm not limited by unknown sparsity, but also optimize the performance of SAMP algorithm. The simulation results show that the algorithm has the advantages in computing speed and reconstruction accuracy when estimating the space-ground mixed channel.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.53

【相似文献】