海杂波背景下运动目标自适应检测方法研究

发布时间：2018-01-16 04:18

本文关键词：海杂波背景下运动目标自适应检测方法研究　出处：《西安电子科技大学》2014年硕士论文　论文类型：学位论文

【摘要】：海杂波背景下的目标检测是雷达目标检测领域的一个重要课题。由于海杂波的复杂特性,海杂波背景下的目标检测也是雷达目标检测领域的一大难题。国内外许多专家和学者一直致力于海杂波特性和海杂波背景下目标检测技术的研究和探索,形成了一定的理论体系并取得了大量的研究成果。本文对海杂波背景下运动目标的自适应检测方法进行了研究,并基于现有方法,提出了适用于非均匀海杂波环境的基于功率中值和归一化协方差估计的自适应检测方法。在中等累积时间条件下,为了检测低速运动弱目标,提出了一种基于中值归一化多普勒功率谱的积分方法。本文首先对海杂波的物理组成机理、统计特性和统计模型进行了全面回顾,然后详细介绍了三种现有的自适应检测方法,它们都高度依赖于杂波向量的协方差矩阵估计,于是进一步回顾了几种常用的杂波协方差矩阵估计方法。对于非均匀的海杂波环境,提出了一种基于功率中值和归一化的协方差矩阵估计方法,该方法首先将参考单元的杂波功率做了归一化处理,然后利用归一化后的参考单元数据来估计归一化协方差矩阵,利用参考单元的功率中值作为待检测单元杂波功率的估计值,功率中值估计和归一化协方差矩阵估计的乘积就形成了待检测单元杂波协方差矩阵的有效估计,这种新的协方差矩阵估计方法能有效抑制海杂波在距离上的功率起伏产生的影响。利用这种新的杂波协方差矩阵估计代替自适应检测器中的杂波协方差矩阵估计就得到了基于功率中值和归一化协方差估计的自适应检测方法,它可以改善检测器在非均匀海杂波环境下的检测性能。利用实测海杂波数据对提出方法的性能改善进行验证,实验结果表明:在非均匀海杂波环境下,采用基于功率中值和归一化协方差矩阵估计的自适应检测方法比采用传统样本协方差矩阵估计的自适应方法有明显的性能优势。接着,本文介绍了海杂波时间序列功率谱估计的基本概念和方法,并分析了海杂波多普勒功率谱特性,结果表明海杂波在每个多普勒通道的功率分布是非均匀的,这就意味着不同多普勒通道的功率值具有不同的均值和方差。另外,在几十毫秒到几秒范围的中等累积时间条件下,运动目标的多普勒偏移通常会分布在几个相邻的多普勒通道中。待检测单元多普勒功率谱与参考距离单元多普勒功率谱中值的比值形成了待检测单元的中值归一化多普勒功率谱,针对低速运动弱目标,提出了一种对中值归一化多普勒功率谱沿多普勒通道的积分方法。利用实测海杂波数据对提出方法进行了检验,结果表明该方法比现有的基于分形的方法有更好的检测性能。
[Abstract]:Target detection in the background of sea clutter is an important subject in radar target detection field, because of the complex characteristics of sea clutter. Target detection in sea clutter background is also a difficult problem in the field of radar target detection. Many experts and scholars at home and abroad have been devoted to Yu Hai clutter characteristics and sea clutter background target detection technology research and exploration. A certain theoretical system has been formed and a lot of research results have been made. In this paper, adaptive detection methods for moving targets in sea clutter background are studied, and based on existing methods. An adaptive detection method based on power median and normalized covariance estimation for heterogeneous sea clutter environment is proposed in order to detect low speed moving weak targets under medium cumulative time. An integral method based on median normalized Doppler power spectrum is proposed. Firstly, the physical composition mechanism, statistical characteristics and statistical model of sea clutter are reviewed. Then three existing adaptive detection methods are introduced in detail, which are highly dependent on the covariance matrix estimation of clutter vectors. Then, several common methods of estimating clutter covariance matrix are reviewed. For non-uniform sea clutter environment, a covariance matrix estimation method based on power median and normalization is proposed. Firstly, the clutter power of the reference unit is normalized, and then the normalized covariance matrix is estimated by the normalized reference unit data. The power median of the reference unit is used as the estimated value of the clutter power of the unit to be detected, and the product of the median power estimation and the normalized covariance matrix estimation is used to form the effective estimation of the clutter covariance matrix of the unit to be detected. This new covariance matrix estimation method can effectively suppress the influence of the power fluctuation of sea clutter on the distance. This new clutter covariance matrix estimation is used to replace the clutter covariance matrix estimation in adaptive detector. An adaptive detection method based on power median and normalized covariance estimation is obtained. It can improve the detection performance of the detector in the non-uniform sea clutter environment. The performance improvement of the proposed method is verified by the measured sea clutter data. The experimental results show that: in the non-uniform sea clutter environment. The adaptive detection method based on the power median and normalized covariance matrix estimation has obvious performance advantages over the traditional sample covariance matrix estimation method. In this paper, the basic concept and method of power spectrum estimation of sea clutter time series are introduced, and the characteristics of Doppler power spectrum of sea clutter are analyzed. The results show that the power distribution of sea clutter in each Doppler channel is non-uniform. This means that the power values of different Doppler channels have different mean and variance. In addition, in the medium cumulative time range of tens of milliseconds to several seconds. The Doppler shift of moving target is usually distributed in several adjacent Doppler channels. The ratio of the Doppler power spectrum of the unit to the reference unit forms the median normalization of the unit to be detected. Doppler power spectrum. A method of integrating the median normalized Doppler power spectrum along the Doppler channel is proposed for the weak target moving at low speed. The proposed method is verified by the measured sea clutter data. The results show that this method has better detection performance than the existing fractal based methods.
【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】：TN957.51

【相似文献】