相干信号的序贯分组波达方向估计算法研究

发布时间：2018-04-02 11:14

本文选题：阵列信号处理　切入点：波达方向估计　出处：《中国科学技术大学》2017年硕士论文

【摘要】：波达方向(DOA)估计是阵列信号处理的一个重要研究分支,由于其在雷达、通信、声呐、会议系统等多种领域的广泛应用,使得波达方向估计在过去的半个多世纪有着快速的发展。在现实环境中,由于山脉、城市建筑反射产生的多径效应和军事上的人为电磁干扰,使得相干信号大量存在,传统的用于解相干的算法是将所有相干信号放在一起处理,阵列孔径损失较大,这会很大程度限制阵列能处理的相干信号数量和DOA估计精度。此外,实际应用中的阵列天线不可避免的存在各种误差,这会导致DOA估计性能严重下降。因此,研究有误差条件下的DOA估计算法不仅具有重要的理论意义,而且还具有很重要的实用价值。本文通过研究均匀线阵下的相干信号分组波达方向估计,提出了一些更为有效的算法,具体的工作如下:1.提出了基于前向空间平滑的相干信号序贯分组波达方向估计算法,根据组内相干信号的个数从少到多依次分组估计。首先估计信号数最少的那组相干信号的DOA,并估计这组相干信号的衰落系数,利用斜投影算子将这组相干信号的成分从阵列协方差矩阵中去除,然后估计信号数第二少的那组相干信号,直到所有组相干信号估计完为止。此外,进一步研究了基于前后向空间平滑算法的分组提升DOA估计精度的改进算法,通过斜投影算子将每组相干信号的成分从阵列协方差矩阵中单独提取出来,利用更多子阵的前后向空间平滑进一步提升每组相干信号的估计精度。通过仿真实验验证提出的算法在阵元数少、信噪比低、快拍数少、信号数多的情形下具有较高的DOA估计性能。2.提出了互耦误差下的相干信号序贯分组波达方向估计算法。利用均匀线阵的特点以及互耦误差与间距的关系,可将阵列两端的阵元看作辅助阵元,中间子阵具有相同的互耦环境,通过中间子阵将互耦系数等效到衰落系数,这样可以忽略互耦的影响。因此,可将基于前向空间平滑的相干信号序贯分组波达方向估计算法用于中间子阵,实现互耦误差下的相干信号序贯分组DOA估计。3.提出了互耦误差下的独立信号和相干信号波达方向估计算法。利用独立信号不需要空间平滑就与噪声子空间正交的特点,通过中间子阵将互耦系数等效到信号能量上,先估计出独立信号的DOA,利用已经估计出的DOA进一步估计出互耦系数,然后将独立信号的成分从阵列协方差矩阵去除并补偿互耦系数后,序贯分组估计相干信号的DOA。
[Abstract]:DOA estimation is an important branch of array signal processing, because of its wide application in radar, communication, sonar, conference system and other fields. The DOA estimation has developed rapidly in the past half century. In the real world, because of the multipath effect caused by the reflection of mountains, urban buildings and the military artificial electromagnetic interference, the coherent signals exist in large quantities. The traditional algorithm for decoherence is to process all coherent signals together, and the aperture loss of the array is large, which will greatly limit the number of coherent signals and the DOA estimation accuracy that the array can process. In practical applications, there are inevitable errors in array antennas, which will lead to a serious deterioration of DOA estimation performance. Therefore, it is not only of theoretical significance to study the DOA estimation algorithm under the condition of errors. It is also of great practical value. In this paper, some more effective algorithms are proposed by studying the DOA estimation of coherent signals under uniform linear array. The main work is as follows: 1. A direction of arrival (DOA) estimation algorithm for coherent signals based on forward spatial smoothing is proposed. The number of coherent signals in a group is estimated from small to many. First, the load of the group of coherent signals with the least number of signals is estimated, and the fading coefficients of the coherent signals are estimated. The component of the coherent signal is removed from the array covariance matrix by oblique projection operator, and the second least coherent signal is estimated until all the coherent signals are estimated. An improved algorithm for improving the accuracy of DOA estimation based on the forward and backward spatial smoothing algorithm is further studied. The components of each coherent signal are extracted from the array covariance matrix by oblique projection operator. The estimation accuracy of each group of coherent signals is further improved by using forward and backward spatial smoothing of more subarrays. Simulation results show that the proposed algorithm has less array elements, lower signal-to-noise ratio and fewer rapid-beat numbers. When the number of signals is large, the performance of DOA estimation is higher. 2. An algorithm for estimating the direction of arrival of coherent signals by sequential grouping with mutual coupling error is proposed. The characteristics of uniform linear array and the relationship between mutual coupling error and spacing are used. The element at both ends of the array can be regarded as the auxiliary element, and the intermediate subarray has the same mutual coupling environment. The mutual coupling coefficient can be equivalent to the fading coefficient by the intermediate subarray, so the influence of mutual coupling can be ignored. The direction of arrival (DOA) estimation algorithm of coherent signals based on forward spatial smoothing can be used in intermediate subarrays. The coherent signal sequential block DOA estimation under mutual coupling error is realized. A direction of arrival estimation algorithm for independent signal and coherent signal under mutual coupling error is proposed. The independent signal is orthogonal to the noise subspace without spatial smoothing. The mutual coupling coefficient is equivalent to the signal energy through the intermediate submatrix, the load of the independent signal is estimated first, and the mutual coupling coefficient is further estimated by using the estimated DOA. Then the component of the independent signal is removed from the array covariance matrix and the mutual coupling coefficient is compensated, then the DOA of the coherent signal is estimated by sequential grouping.
【学位授予单位】：中国科学技术大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.7

【相似文献】