α稳定分布噪声背景极化阵列信号DOA参数估计方法研究

发布时间：2018-01-13 15:25

本文关键词：α稳定分布噪声背景极化阵列信号DOA参数估计方法研究　出处：《吉林大学》2017年硕士论文　论文类型：学位论文

【摘要】：波达方向(DOA)估计问题是信号处理领域的一个重要研究内容。研究初期主要是针对传统的标量传感器阵列信号的DOA估计。传统的标量传感器阵列具有一个明显的缺点,它的阵元只能获得3个电场分量的信息,而忽视了信号所包含的极化信息。电磁矢量传感器由中心位置重合的三个相互正交的电偶极子和三个相互正交的磁偶极子构成。随着研究的深入,学者们发现极化阵列可以获得更加完备的信号信息,这使得基于极化阵列信号的DOA估计问题得到了热切关注。但是当时的大量研究都是基于高斯噪声的假设前提下。虽然高斯分布具有不容置疑的研究意义,然而实际环境中只有极少数的随机过程可以严格地满足高斯分布的定义与要求。在实际环境中,人们发现在其测量环境中不仅存在着高斯噪声,而且还大量的存在着一种非高斯的冲击噪声,即a噪声。由于α不存在有限的二阶矩,所以传统的用于处理高斯噪声的信号处理方法在处理a噪声时无法得到很好的效果甚至严重失效。为了对α噪声进行处理,人们提出了基于分数低阶矩的一系列阵列信号DOA估计方法,这一类方法无法对有色a噪声起到抑制作用,并且当遇到与信号频率不同的干扰信号时其DOA估计精度就会有所下降。针对以上一系列的问题,本文通过对分数低阶循环相关方法进行了大量研究,分别结合旋转不变技术(ESPRIT)和总体最小二乘旋转不变技术(TLS-ESPRIT),提出了分数低阶循环相关旋转不变技术(FLOCC-ESPRIT)以及分数低阶循环相关总体最小二乘旋转不变技术(FLOCC-TLS-ESPRIT)方法。本文方法是首先通过分数低阶循环相关的方法对α噪声和高斯噪声进行抑制,并同时对与信号具有不同循环频率的干扰信号进行抑制。再通过经典的DOA估计方法ESPRIT方法和TLS-ESPRIT方法对信号进行DOA和极化参数的估计。文中分别将上述方法应用于单电磁矢量传感器,线性极化阵列以及任意结构极化阵列。最后经过MATLAB仿真实验验证,本文方法具有很好的噪声抑制效果以及很好的DOA估计性能。
[Abstract]:DOA). The estimation problem is an important research content in the field of signal processing. At the beginning of the research, it mainly aims at the DOA estimation of the traditional scalar sensor array signal. The traditional scalar sensor array has an obvious disadvantage. Its array element can only obtain the information of three electric field components. The electromagnetic vector sensor consists of three orthogonal electric dipoles and three orthogonal magnetic dipoles. Researchers have found that polarization arrays can obtain more complete signal information. This makes the problem of DOA estimation based on polarimetric array signal get much attention. However, a lot of researches were based on the assumption of Gao Si noise at that time. Although the Gao Si distribution has no doubt about the significance of the research. However, only a few stochastic processes can strictly meet the definition and requirements of Gaussian distribution. In the actual environment, it is found that there is not only Gaussian noise in the measurement environment. There is also a large number of non-#china_person0# impact noise, that is, a noise, because 伪 does not have a finite second order moment. Therefore, the traditional signal processing method used to deal with Gao Si noise can not get a good effect or even serious failure when dealing with a noise. In order to deal with 伪 noise. A series of array signal DOA estimation methods based on fractional lower moments are proposed, which can not suppress the colored a noise. And the accuracy of DOA estimation will decrease when the interference signal is different from the signal frequency. In view of the above series of problems, this paper has done a lot of research on the fractional low order cyclic correlation method. Combined with rotational invariant technique (Esprit) and total least-squares rotation invariant technique (TLS-ESPRIT), respectively. The fractional low order cyclic correlation rotation invariant technique (FLOCC-Esprit) and the fractional low order cyclic correlation total least squares rotation invariant technique (FLOCC-Esprit) are presented. The method of FLOCC-TLS-ESPRIT.The first step is to suppress 伪 noise and Gao Si noise by fractional low order cyclic correlation (FLOCC-TLS-Esprit) method. At the same time, the interference signal with different cycle frequency is suppressed. Then the DOA and polarization of the signal are carried out by the classical DOA estimation method ESPRIT and the TLS-ESPRIT method. The method is applied to single electromagnetic vector sensor. Linear polarization array and arbitrary polarization array. Finally, the MATLAB simulation results show that the proposed method has good noise suppression effect and good DOA estimation performance.
【学位授予单位】：吉林大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN911.7

【相似文献】