【作者】 李鹏；

【导师】 饶鲜；

【作者基本信息】 西安电子科技大学 ， 电路与系统， 2017， 硕士

【摘要】 电子侦察是电子战重要的组成部分。目标的定位跟踪是电子侦察的核心。伴随着科学技术的不断进步,电磁环境变得日益复杂。对目标定位跟踪的范畴已经从针对单一目标的定位跟踪发展为针对多目标的定位跟踪。本文分别从定位和跟踪两个方面研究了多目标的定位跟踪问题。在无源测向定位中,测向交叉定位法是使用较多的一种定位方法。首先分析了测向交叉定位的基本原理,然后重点研究了测向交叉定位法在多目标定位中的应用。与此同时提出了一种改进Hough变换法来去除多目标测向交叉定位中的虚假定位点(也称为鬼点)。多目标跟踪主要包括目标航迹的起始,目标的跟踪滤波以及航迹的终结三个部分。本文重点研究了多目标的航迹起始和多目标的跟踪滤波。数据关联技术是贯穿多目标跟踪过程的一项关键技术。在经典的数据关联算法中,最近邻域法计算量最小,在工程上对多目标的数据关联有很大的优势。但其跟踪性能较差,很容易发生误跟或失跟的情况。所以本文提出了一种优化的最近邻数据关联算法来弥补这一缺陷。因为无源定位数据具有不连续性和随机性的特点,经典的航迹起始算法不能用于无源定位系统的航迹起始。搜索法是一种可以适用于无源定位系统的多目标航迹起始算法。卡尔曼滤波是最经典的线性滤波算法。扩展卡尔曼滤波是最经典的非线性滤波算法。这两种滤波算法经常用来对单目标的航迹进行跟踪滤波,而本文的重点是将这两种滤波算法应用到多目标的跟踪滤波中。粒子滤波算法是近些年兴起的一种滤波算法,尚处在理论研究阶段。本文对这种算法的理论进行了分析。更多还原

【Abstract】 Electronic reconnaissance is an important part of electronic warfare.The location and track of the target is the core of electronic reconnaissance.Along with the continuous progress of science and technology,the electromagnetic environment is becoming more and more complicated.The scope of target location and track has evolved from single target to multi-target.The location and track problems of multi-target are studied in this paper from two aspects: location and track.In the multi-target passive DOA location,the direction-finding cross location method is one of the most commonly used methods.Firstly,the basic principle of the direction-finding cross location is analyzed,and then the application of the direction-finding cross location in multi-target location is studied emphatically.At the same time,an improved Hough transform method is proposed to remove the false location points(also called ghosts)in the multi-target location.Multi-target track mainly includes the target track initiation,the target tracking fitering and the target track termination.This paper focuses on the multi-target track initiation and the multi-target tracking filtering.Data association technology is a key technology in the process of multi-target track.In the classical data association algorithm,the nearest neighbor method has a great advantage with less compution in the engineering multi-target data association.But its tracking accuracy is very poor and prones to misconduct.An optimized nearest neighbor data association algorithm is proposed to compensate for this flaw.Because passive location data has the characteristics of discontinuity and randomness,the classical track initiation algorithm can not be used to the passive location system.The search method is a multi-target track initiation algorithm which can be applied to the passive location system.Kalman filter is the most classical linear filtering algorithm.The extended Kalman filter is the most classical nonlinear filtering algorithm.These two filtering algorithms are often used to track the single target,while this paper focuses on how to apply these two filtering algorithm to multi-target tracking filtering.Particle filter algorithm is a new kind of filtering algorithm which emerged in recent years and is still at the stage of theoretical research.The theory of this algorithm is analyzed in this paper.更多还原