基于一维距离像的弹道目标中段微动特征提取
发布时间：2018-12-31 09:23
【摘要】：弹道导弹目标识别技术是弹道导弹防御(BMD)系统的重要研究内容,从攻击目标群中准确识别出弹头目标,是成功实现导弹拦截的前提。弹道导弹中段飞行的时间最长,是真假弹头识别的关键阶段。弹道中段目标的高分辨一维距离像(HRRP)序列中包含了目标的微动特征,利用这些微动特征可以有效的区分弹头和诱饵。本文在分析导弹目标微动特性的基础上,基于弹道中段目标HRRP序列分别从三个不同途径提取出了目标的微动特征参数,主要研究内容如下:1、建立了弹道中段目标的微动模型,包括锥体弹头的进动、章动和锥体诱饵的摆动、翻滚,然后通过分析弹道目标的微动特性,推导了不同微动状态下的雷达微多普勒表达式,并通过比较各种微动状态下微多普勒存在的差异,验证了微动特征在区分弹头和诱饵上的有效性和可行性。2、研究了基于二维图形检测的目标微动参数估计方法。该方法一方面直接利用HRRP序列加权积累得到目标进动周期的估计曲线,并分别利用MUSIC算法和非线性最小二乘的Gauss-Newton方法估计了目标的进动频率,另一方面基于广义Radon变换(GRT)的曲线检测能力估计了目标的进动角,与传统的RCS多项式拟合方法相比,改善了估计精度,且无需事先得知目标的物理结构。3、针对二维图形检测的计算复杂度会随着检测曲线参数个数和精度要求的上升而大幅度增加的缺点,研究了基于各散射点RLOS起伏特性曲线的弹道目标中段微动参数估计方法。该方法利用点迹-航迹配对方法提取出目标各散射点的RLOS起伏特性曲线,一方面从散射点的RLOS起伏特性包含的频率成分中提取出了目标的进动频率,另一方面通过提取锥体目标顶点散射点的RLOS投影距离极值并根据几何关系估计了弹道中段进动目标的进动角。4、为了提高参数估计的精度,研究了基于散射中心位置差序列的进动参数估计方法。该方法利用锥体弹头底面边缘处始终迎着RLOS的散射点与锥体顶点散射点在RLOS方向上投影的位置差序列与进动参数之间的函数关系,估计出弹头的进动频率和进动角。与利用较大时间间隔内雷达视线角的变化估计进动角的方法相比,该方法的估计精度和稳定性都有所提高。
[Abstract]:Ballistic missile target recognition technology is an important research content of ballistic missile defense (BMD) system. Accurately identifying warhead target from attack target group is the premise of successful missile interception. The longest flight time in the middle of ballistic missile is the key stage of real and false warhead identification. The high-resolution one-dimensional range profile (HRRP) sequence of a target in the middle of a trajectory contains the fretting characteristics of the target, which can be used to effectively distinguish the warhead from the decoy. On the basis of analyzing the fretting characteristics of missile target, the fretting characteristic parameters of the missile target are extracted from three different ways based on the HRRP sequence of the mid-ballistic target. The main research contents are as follows: 1. The fretting model of the target in the middle part of the trajectory is established, including the precession of the pyramidal warhead, the nutation and the swinging and rolling of the pyramidal decoy. Then, by analyzing the fretting characteristics of the ballistic target, the expression of radar microDoppler in different fretting states is derived. By comparing the differences of micro-Doppler in various fretting states, the validity and feasibility of fretting feature in distinguishing warhead from decoy are verified. 2. The method of fretting parameter estimation based on two-dimensional graph detection is studied. On the one hand, the estimation curve of the precession period of the target is obtained by the weighted accumulation of HRRP sequence, and the precession frequency of the target is estimated by using the MUSIC algorithm and the nonlinear least square Gauss-Newton method, respectively. On the other hand, based on the curve detection ability of generalized Radon transform (GRT), the precession angle of the target is estimated. Compared with the traditional RCS polynomial fitting method, the estimation accuracy is improved, and the physical structure of the target is not known in advance. Aiming at the shortcoming that the computational complexity of two-dimensional graph detection will increase greatly with the increase of the number and precision of the parameters of the detection curve, a method for estimating the fretting parameters of ballistic targets based on the RLOS fluctuation characteristic curve of each scattering point is studied. In this method, the RLOS fluctuation characteristic curve of each scattering point is extracted by point-track matching method. On the one hand, the precession frequency of the target is extracted from the frequency components contained in the RLOS fluctuation characteristic of the scattering point. On the other hand, by extracting the RLOS projection distance extremum of the scattering point at the vertex of the cone target and estimating the precession angle of the precession target in the middle part of the trajectory according to the geometric relation, the precession angle of the precession target in the middle part of the trajectory is estimated. The precession parameter estimation method based on scattering center position difference sequence is studied. In this method, the precession frequency and precession angle of the warhead are estimated by using the function relationship between the position difference sequence and the precession parameters of the RLOS scattering point and the cone vertex scattering point projection in the RLOS direction at the bottom edge of the cone warhead. Compared with the method of estimating precession angle by using the variation of radar line of sight angle in large time interval, the accuracy and stability of this method are improved.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TJ761.3;TN957.51
本文编号：2396374
[Abstract]:Ballistic missile target recognition technology is an important research content of ballistic missile defense (BMD) system. Accurately identifying warhead target from attack target group is the premise of successful missile interception. The longest flight time in the middle of ballistic missile is the key stage of real and false warhead identification. The high-resolution one-dimensional range profile (HRRP) sequence of a target in the middle of a trajectory contains the fretting characteristics of the target, which can be used to effectively distinguish the warhead from the decoy. On the basis of analyzing the fretting characteristics of missile target, the fretting characteristic parameters of the missile target are extracted from three different ways based on the HRRP sequence of the mid-ballistic target. The main research contents are as follows: 1. The fretting model of the target in the middle part of the trajectory is established, including the precession of the pyramidal warhead, the nutation and the swinging and rolling of the pyramidal decoy. Then, by analyzing the fretting characteristics of the ballistic target, the expression of radar microDoppler in different fretting states is derived. By comparing the differences of micro-Doppler in various fretting states, the validity and feasibility of fretting feature in distinguishing warhead from decoy are verified. 2. The method of fretting parameter estimation based on two-dimensional graph detection is studied. On the one hand, the estimation curve of the precession period of the target is obtained by the weighted accumulation of HRRP sequence, and the precession frequency of the target is estimated by using the MUSIC algorithm and the nonlinear least square Gauss-Newton method, respectively. On the other hand, based on the curve detection ability of generalized Radon transform (GRT), the precession angle of the target is estimated. Compared with the traditional RCS polynomial fitting method, the estimation accuracy is improved, and the physical structure of the target is not known in advance. Aiming at the shortcoming that the computational complexity of two-dimensional graph detection will increase greatly with the increase of the number and precision of the parameters of the detection curve, a method for estimating the fretting parameters of ballistic targets based on the RLOS fluctuation characteristic curve of each scattering point is studied. In this method, the RLOS fluctuation characteristic curve of each scattering point is extracted by point-track matching method. On the one hand, the precession frequency of the target is extracted from the frequency components contained in the RLOS fluctuation characteristic of the scattering point. On the other hand, by extracting the RLOS projection distance extremum of the scattering point at the vertex of the cone target and estimating the precession angle of the precession target in the middle part of the trajectory according to the geometric relation, the precession angle of the precession target in the middle part of the trajectory is estimated. The precession parameter estimation method based on scattering center position difference sequence is studied. In this method, the precession frequency and precession angle of the warhead are estimated by using the function relationship between the position difference sequence and the precession parameters of the RLOS scattering point and the cone vertex scattering point projection in the RLOS direction at the bottom edge of the cone warhead. Compared with the method of estimating precession angle by using the variation of radar line of sight angle in large time interval, the accuracy and stability of this method are improved.
【学位授予单位】：电子科技大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TJ761.3;TN957.51
【相似文献】
相关期刊论文 前10条
1 汤军;孙伟;;弹道目标跟踪的自适应多维分配相关算法[J];弹道学报;2011年02期
2 杨有春;童宁宁;冯存前;贺泽维;贺思三;;弹道目标中段平动补偿与微多普勒提取[J];宇航学报;2011年10期
3 刘也;余安喜;朱炬波;唐歌实;;弹道目标实时跟踪中的滤波方法综述[J];宇航学报;2013年11期
4 巫春玲;韩崇昭;;用于弹道目标跟踪的有限差分扩展卡尔曼滤波算法[J];西安交通大学学报;2008年02期
5 徐长爱;李尚生;殷勇;刘军;;基于粒子滤波的弹道目标跟踪[J];电光与控制;2008年10期
6 饶彬;屈龙海;肖顺平;王雪松;;基于时间序列分析的弹道目标进动周期提取[J];电波科学学报;2011年02期
7 雷腾;刘进忙;李松;胡国平;;弹道目标进动周期特征提取新方法[J];计算机工程与应用;2013年01期
8 俞建国;刘梅;包玖红;姚璐;;基于星光测量天基传感器实时定标在弹道目标跟踪应用[J];电子与信息学报;2013年04期
9 王军;弹道目标空间反识别措施发展态势浅析[J];系统工程与电子技术;2002年07期
10 罗贤明;陈忠宽;刘捷;程阳;;弹道目标RCS计算与分析[J];空军雷达学院学报;2011年05期
相关博士学位论文 前1条
1 刘也;弹道目标实时跟踪的稳健高精度融合滤波方法[D];国防科学技术大学;2011年
相关硕士学位论文 前8条
1 杨建辉;基于一维距离像序列的弹道目标微动参数估计方法研究[D];电子科技大学;2012年
2 周旋;弹道目标轨道确定方法研究[D];西安电子科技大学;2014年
3 施岩龙;非线性滤波技术在弹道目标跟踪中的应用[D];南京理工大学;2012年
4 袁仕继;基于宽带信息的中段弹道目标特征提取方法研究[D];国防科学技术大学;2009年
5 宗志伟;再入弹道目标跟踪与质阻比识别方法研究[D];国防科学技术大学;2010年
6 高铭江;弹道目标运动参数估计方法研究及软件设计[D];西安电子科技大学;2014年
7 明达;基于一维距离像的弹道目标中段微动特征提取[D];电子科技大学;2014年
8 黄双宁;弹道目标信息处理技术研究[D];哈尔滨工业大学;2009年
，本文编号：2396374
本文链接：https://www.wllwen.com/kejilunwen/wltx/2396374.html
