非圆信号参数估计方法研究

发布时间：2018-06-15 19:16

本文选题：非圆信号 + 时差估计　；参考：《电子科技大学》2014年博士论文

【摘要】：非圆信号是现代无线通信中常见的一种信号，例如AM、MASK、UQPSK、BPSK、MSK、GMSK和OQPSK信号。本文研究非圆信号的参数估计技术，包括非圆信号的TDOA估计、基于非相参阵列的非圆信号DOA估计、以及非圆信号的稳健Capon波束形成。与圆信号情况下只利用信号的相关统计信息相比，，非圆信号参数估计研究的关键在于如何联合利用信号的相关和共轭相关统计信息来提高参数估计的性能。本文的主要工作概括如下： 1.在高斯非圆复信号和高斯圆噪声的模型下推导了TDOA估计的最大似然算法和CRB。与互相关和共轭互相关算法相比，最大似然算法更全面的利用了非圆信号的二阶统计信息，因此在非圆信号情况下有更好的估计性能。同时，针对非圆信号提出了一种改进传统循环算法的方案，新方案可以同时利用信号的循环相关和共轭循环相关信息。 2.从信息论的角度出发，在高斯和拉普拉斯两种实信号假设下对比了基于联合熵和互信息两个准则的实信号TDOA估计算法。然后，在非圆高斯复信号和圆高斯复噪声的模型下推导了基于互信息准则的TDOA估计算法。 3.研究了基于非相参阵列的非圆信号渐进最优二阶DOA估计算法。首先，在非圆高斯复信号和圆高斯噪声的假设下推导了非相参阵列DOA估计的最大似然算法和闭式CRB。接着，提出了改进的广义最小二乘算法，此算法对高斯或非高斯圆信号都是渐进最小方差的二阶算法。随后，提出了一种非圆信号的渐进最小方差二阶算法。 4.研究了非相参阵列的MUSIC类DOA估计算法，提出了只使用第一协方差矩阵的子阵加权MUSIC算法（w-MUSIC），并把w-MUSIC算法扩展到非圆信号。同时，对比分析了w-MUSIC算法和文献[85]中MUSIC算法的渐进估计方差。 5.针对SOI和干扰信号都可能是非圆信号的情况提出了稳健的自适应波束形成方法NC-RCB。NC-RCB不但对阵列方向向量误差和协方差矩阵估计误差是稳健的，同时对SOI非圆系数的估计误差也是稳健的。在SOI是非圆信号的情况下，NC-RCB与只利用第一协方差矩阵的稳健Capon波束形成方法相比具有显著的性能提升。
[Abstract]:Non-circular signal is a common signal in modern wireless communication, such as AMMASK / UQPSK / BPSK / MSK / GMSK and OQPSK signal. This paper studies the parameter estimation techniques of non-circular signals, including TDOA estimation of non-circular signals, DOA estimation of non-circular signals based on non-coherent arrays, and robust Capon beamforming of non-circular signals. Compared with only using the correlation statistics of the signal in the case of circular signal, the key of the parameter estimation of non-circular signal is how to combine the correlation and conjugate correlation statistics of the signal to improve the performance of the parameter estimation. The main work of this paper is summarized as follows: 1. Based on the model of Gao Si noncircular complex signal and Gao Si circular noise, the maximum likelihood algorithm and Gao Si estimation algorithm are derived. Compared with the cross-correlation and conjugate cross-correlation algorithms, the maximum likelihood algorithm makes full use of the second-order statistical information of the non-circular signal, so it has better estimation performance in the case of non-circular signal. At the same time, an improved algorithm for non-circular signals is proposed, which can make use of the cyclic correlation information and conjugate cyclic correlation information simultaneously. 2. From the viewpoint of information theory, the real signal estimation algorithms based on joint entropy and mutual information are compared under the assumption of Gao Si and Laplace. Then, under the models of non-circular Gao Si complex signal and circular Gao Si complex noise, the estimation algorithm based on mutual information criterion is derived. An asymptotic optimal second order DOA estimation algorithm for noncircular signals based on noncoherent arrays is studied. Firstly, under the assumption of non-circular Gao Si complex signal and circular Gao Si noise, the maximum likelihood algorithm and closed Gao Si estimation algorithm for noncoherent array DOA estimation are derived. Then, an improved generalized least squares algorithm is proposed, which is a second-order algorithm with asymptotic minimum variance for both Gao Si and non-Gao Si circular signals. Then, an asymptotic minimum variance second order algorithm for noncircular signals is proposed. 4. The music class DOA estimation algorithm for noncoherent arrays is studied. A subarray weighted music algorithm using only the first covariance matrix is proposed, and the w-MUSIC algorithm is extended to non-circular signals. At the same time, the asymptotic estimated variance of the w-MUSIC algorithm and the music algorithm in literature [85] are compared and analyzed. In this paper, a robust adaptive beamforming method NC-RCB.NC-RCB is proposed for the case that both SOI and interference signals may be non-circular signals. NC-RCB.NC-RCB is not only robust for array direction vector errors and covariance matrix estimation errors. At the same time, the estimation error of SOI noncircular coefficient is also robust. In the case of SOI non-circular signals, NC-RCB has a significant performance improvement compared with the robust Capon beamforming method using only the first covariance matrix.
【学位授予单位】：电子科技大学
【学位级别】：博士
【学位授予年份】：2014
【分类号】：TN911.23

【共引文献】