空时分布源DOA-时延参数联合估计算法研究

发布时间：2018-06-22 22:22

  本文选题：空时分布源 + 无线通信系统　； 参考：《电子科技大学》2015年硕士论文

【摘要】：在复杂的无线通信环境中,用户信号经过折射、绕射以及散射等原因产生多径传播现象导致在空间上、时间上的角度扩展和时延扩展造成码间干扰以及同信道干扰,这些干扰都是影响现代无线移动通信质量的重要原因。在无线移动通信系统中采用空、时联合处理技术可以有效地抑制信号的多径传播、增大系统容量以及提高通信质量等好处,具有重要的理论意义和适用价值。本论文针对上述问题,以无线通信环境多径传播为研究对象,深入地分析了现有信道模型,对空时分布式多径信号参数联合估计算法进行了改进以及提出了新的估计方法,论文的主要内容如下:基于已知分布函数情况下的空时分布式信道模型,分析了空时信道模型噪声间的相干关系以及经过离散傅里叶变换和去卷积后的信道噪声方差形式,发现传统的空时参数联合估计方法在经过去卷积后可使得信道的噪声方差变大从而影响信号子空间,从而导致估计的不准确。针对上述问题提出了一种改进方法,通过计算空间协方差矩阵和时延协方差矩阵来构建空间传播算子和时延传播算子,利用传播算子和阵列流行向量正交性质可以搜索出空时联合参数信息,然后通过联合传播算子的正交投影来实现空时参数配对。根据均匀线阵相邻元素间相位上保持旋转不变特性,计算空间算子和时延旋转算子来估计空间和时延参数,该方法无需谱峰搜索就能够有效的估计出中心DOA和中心时延。与现有的最大似然准则、谱估计以及子空间的空时分布式多径信号参数联合估计方法相比,本文提出的改进子空间分解方法能够有效地避免协方差矩阵特征值分解以及多维参数估计中优化问题,从而降低算法复杂度,而且具有较好的算法性能。分析了空时分布源多径簇信号在信道中的稀疏特性,提出了基于稀疏重构的相干空时分布式多径簇信号的参数联合估计算法,该算法在未知空时分布式多径簇信号的角度密度函数和时延密度函数的情况下能够有效地估计出中心DOA和中心时延,在知道其角度和时延的密度函数的分布扩展形式下,则可以利用稀疏重构方法估计出其角度扩展和时延扩展,具有估计精度高、分辨率好以及对信号的分布特性不敏感等优点。
[Abstract]:In the complex wireless communication environment, the multipath propagation of user signals caused by refraction, diffraction and scattering causes intersymbol interference and cochannel interference in space, time angle spread and time delay spread. These interference are the important reasons that affect the quality of modern wireless mobile communication. The use of space-time joint processing technology in wireless mobile communication systems can effectively suppress the multipath propagation of signals, increase the system capacity and improve the communication quality. It has important theoretical significance and applicable value. Aiming at the above problems, this paper takes multipath propagation in wireless communication environment as the research object, deeply analyzes the existing channel models, improves the joint estimation algorithm of space-time distributed multipath signal parameters, and proposes a new estimation method. The main contents of this paper are as follows: based on the space-time distributed channel model with known distribution function, the coherent relation between the noise of space-time channel model and the variance of channel noise after discrete Fourier transform and deconvolution are analyzed. It is found that the traditional joint space-time parameter estimation method can cause the channel noise variance to increase after deconvolution, thus affecting the signal subspace, which leads to the inaccuracy of the estimation. In order to solve the above problems, an improved method is proposed to construct spatial propagation operator and delay propagation operator by computing spatial covariance matrix and delay covariance matrix. Space-time joint parameter information can be searched by orthogonal property of propagation operator and array popular vector, and then space-time parameter pairing can be realized by orthogonal projection of joint propagator. The spatial and delay parameters are estimated by calculating the spatial operator and the time-delay rotation operator according to the rotation invariant property of the phase between adjacent elements of the uniform linear array. This method can effectively estimate the central DOA and the central delay without the spectral peak search. Compared with the existing maximum likelihood criterion, spectral estimation and space-time distributed multipath signal parameter estimation in subspace, The improved subspace decomposition method proposed in this paper can effectively avoid the eigenvalue decomposition of covariance matrix and the optimization problem in multidimensional parameter estimation, thus reducing the complexity of the algorithm and having better algorithm performance. The sparse characteristics of space-time distributed source multipath cluster signals in the channel are analyzed, and a joint parameter estimation algorithm for coherent space-time distributed multipath cluster signals based on sparse reconstruction is proposed. The algorithm can effectively estimate the central DOA and the central delay when the angular density function and the delay density function of the distributed multipath cluster signals are unknown. The sparse reconstruction method can be used to estimate the angular spread and delay spread, which has the advantages of high estimation accuracy, good resolution and insensitivity to the signal distribution characteristics.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TN929.5
，

本文编号：2054434