利用压缩感知提升极化敏感阵列参数估计性能的方法研究

发布时间：2018-12-18 05:15

【摘要】：极化敏感阵列信号处理相比标量天线增加了极化参数估计维度,通过同时利用入射信号的空域和极化域信息,提升了阵列信号处理的输出性能(包括参数估计),所以越来越多地受到阵列信号处理领域研究人员的关注。传统的联合极化—空域谱估计方法利用极化—空域的旋转不变性求解波达方向(俯仰角、方位角)和极化状态(辅极化角、极化相位差)。然而实际中存在着幅相误差,造成旋转不变性难以满足,导致参数估计性能下降甚至完全失效。针对此问题,考虑采用压缩感知在降低硬件复杂度和算法复杂度的同时,提高极化敏感阵列在幅相误差情况下参数估计性能。为此,本文展开了基于完备电磁矢量传感器和三正交偶极子阵列的利用压缩感知提升极化敏感阵列参数估计性能的研究。针对完备电磁矢量传感器参数估计精度不佳的问题,提出一种基于完备电磁矢量和磁场Z轴极化方向平面阵的联合极化—空域VS-OMP(Vector-Sensor Orthogonal Matching Pursuit)压缩感知谱估计方法。首先,对位于坐标轴原点的完备电磁矢量传感器接收的信号,利用时移不变性粗估波达方向和极化状态;其次,以粗估波达方向为中心、给定误差范围为半径建立平面阵的空域局部拼接字典;然后,通过正交匹配跟踪方法实现波达方向精估计。仿真结果验证了该方法的有效性,并且体现了该方法对单信源情形下性能接近于旋转不变子空间(Estimation of Signal Parameters via Rotational Invariance Techniques,ESPRIT)方法,但在强弱多信源下,ESPRIT方法和VS-OMP方法估计精度都不高。针对存在强弱信号并存和存在幅相误差的情况,提出一种基于三正交偶极子阵列的强弱信号分离的稳健压缩感知参数估计方法,首先,对三正交偶极子阵列接收的三个电场分量,通过OMP方法实现X、Y、Z三组偶极子阵列的波达方向估计,数值平均这三组估计结果;其次,采用保形调零方法实现对弱信号的增强和强信号的抑制,并且避免了对极化信息的污染,最后,对归一化的电场矢量解算出强弱目标的极化状态。仿真结果表明与传统方法相比,所提方法在存在幅相误差时提升了极化—空域参数估计性能,在有相干源情况下,无需额外处理即可实现相干源的参数估计。总体而言,相比基于ESPRIT的传统联合极化—空域谱估计方法,在相干源和存在幅相误差的情形下利用压缩感知方法可以提高极化敏感阵列参数估计性能。
[Abstract]:Compared with scalar antennas, polarization-sensitive array signal processing increases the dimension of polarization parameter estimation, and improves the output performance of array signal processing (including parameter estimation) by using the spatial and polarization domain information of incident signal simultaneously. So more and more researchers in the field of array signal processing pay more and more attention. The traditional joint polarization-spatial domain spectral estimation method uses the rotation invariance of polarization-spatial domain to solve the direction of arrival (pitch angle, azimuth) and polarization state (auxiliary polarization angle, polarization phase difference). However, there are amplitude and phase errors in practice, which result in the rotation invariance is difficult to satisfy, and the performance of parameter estimation is degraded or even completely invalid. To solve this problem, compression sensing is considered to reduce the complexity of hardware and algorithm and to improve the performance of parameter estimation of polarization-sensitive arrays in the case of amplitude-phase error. Therefore, based on complete electromagnetic vector sensor and tri-orthogonal dipole array, the parameter estimation performance of polarization-sensitive array based on compression sensing is studied. Aiming at the problem of poor precision of parameter estimation of complete electromagnetic vector sensor, a joint polarization-spatial domain VS-OMP (Vector-Sensor Orthogonal Matching Pursuit) compressed sensing spectrum estimation method) based on complete electromagnetic vector and Z-axis polarization direction plane array of magnetic field is proposed. Firstly, the time shift invariance is used to estimate the direction of arrival (DOA) and polarization state of the signals received by the complete electromagnetic vector sensor at the origin of the coordinate axis. Secondly, taking the direction of arrival as the center and the given error range as the radius, the spatial local splicing dictionary of the plane array is established, and then the precise estimation of the direction of arrival is realized by the orthogonal matching tracking method. The simulation results verify the effectiveness of the proposed method, and show that the performance of the method is close to that of the rotating invariant subspace (Estimation of Signal Parameters via Rotational Invariance Techniques,ESPRIT method in the case of single source, but in the case of strong or weak multi-source, the performance of the proposed method is similar to that of the rotation-invariant subspace method. The estimation accuracy of ESPRIT method and VS-OMP method are not high. In view of the coexistence of strong and weak signals and the existence of amplitude and phase errors, a robust compression sensing parameter estimation method based on the separation of strong and weak signals based on three orthogonal dipole arrays is proposed. For the three electric field components received by the triorthogonal dipole array, the direction of arrival (DOA) estimation of the three sets of dipole arrays is realized by using the OMP method, and the numerical average of these three sets of estimation results are obtained. Secondly, the enhancement of weak signal and the suppression of strong signal are realized by the method of shape preserving zero adjustment, and the pollution of polarization information is avoided. Finally, the polarization state of strong and weak target is calculated by solving the normalized electric field vector solution. The simulation results show that the proposed method improves the performance of polarization-spatial parameter estimation in the presence of amplitude-phase errors. In the case of coherent sources, the parameter estimation of coherent sources can be realized without additional processing. In general, compared with the traditional joint polarization-spatial spectral estimation method based on ESPRIT, the performance of polarization-sensitive array parameter estimation can be improved by using compressed sensing method in the case of coherent source and amplitude and phase error.
【学位授予单位】：西安电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN911.7

【相似文献】