【作者】 李亚超；

【导师】 保铮；

【作者基本信息】 西安电子科技大学 ， 信号与信息处理， 2008， 博士

【摘要】 微波成像雷达具有全天时、全天候、远距离提供高分辨图像的能力。由于它在复杂的环境条件下,可完成常规监视系统(光学、红外成像等)难以胜任的任务,因此,在军事和民用领域受到了越来越广泛的关注和应用。随着海洋和空间资源的不断开发和利用,海域和领空资源监测与安全防御也将变得更为重要,因此,对海面或空间目标进行检测和高分辨逆合成孔径雷达(ISAR)成像,实现目标的监测和识别,将显得尤为重要,具有重大的经济、军事和科研价值。本文主要研究海面或空间目标的运动参数估计和ISAR成像方法。提出了基于相邻相关法的宽带信号的运动参数估计方法;建立了目标复杂运动回波信号模型,分析了目标运动给方位聚焦带来的影响,提出了基于时频分布尺度变换和时间-调频率分布的ISAR成像方法;在已有的技术基础之上,把雷达推广到机载平台上,实现舰船目标的合成孔径雷达高分辨成像。具体的工作概括如下:1、论文的第一章为绪论,主要回顾了目标检测和ISAR成像的发展历程,并介绍了本文的研究背景和主要工作。2、第二章介绍了ISAR成像的基本原理。对常规ISAR成像中所用到的运动补偿技术和方位聚焦方法进行了概述。同时,分析了时频分析在ISAR成像中的应用,介绍了几种常用的ISAR瞬时成像方法。3、第三章提出了一种有效的高机动目标宽带信号参数估计的方法。先通过相邻相关法对目标回波信号进行降阶处理,除去包络弯曲的影响;后利用keystone变换除去单元内剩余包络走动的影响,对每个距离单元进行瞬时相关尺度变换,并沿方位相干积累在低信噪比情况下检测运动目标。利用瞬时相关尺度变换并相干积累所得到的信号峰值的位置估计目标的径向加速度和加加速度的值,同时,在原始回波信号的相邻相关值中补偿掉高次相位,利用方位相干积累得到的信号峰值位置来估计目标的径向速度。4、对检测之后的目标进行高分辨的ISAR成像,除了直接发射宽频带的信号产生目标的高分辨一维距离像之外,还可以通过窄频带的合成技术实现高分辨的ISAR成像。因此,第四章重点介绍了一种适用于ISAR成像的距离合成高分辨技术,根据ISAR成像的特点,分析了目标运动特性对距离高分辨合成及ISAR成像的影响,提出了基于合成距离包络的运动补偿技术和ISAR成像方法。5、第五章提出了一种基于时频分布尺度变换的ISAR成像新方法,在不需要进行参数估计和运动补偿的情况下,能有效解决方位二次相位项造成方位散焦的问题。把包络对齐后的各距离单元数据变换到时频平面内,通过尺度变换,解信号瞬时时间和相关函数延迟量的耦合影响,把方位二次相位项所产生的时频平面内的斜线校正成平行于时间轴的直线,并沿时间轴进行能量积累,减少交叉项的影响,最终对复杂运动目标进行高分辨ISAR成像。6、第六章提出了一种新的基于时间-调频率平面的解线调频Clean搜索(TC-DechirpClean)的瞬时成像方法。有效地解决了目标复杂运动引起的调频率变化所导致的方位聚焦质量下降问题。在方位信号的时间-调频率平面内,利用信号散射点调频率变化率的一维搜索方法得到信号的调频率的变化率和调频率值,同时,在补偿高次相位后做FFT变换,根据幅度最大的位置来确定信号的中心频率和幅度信息,重构有用的回波信号,最终对目标进行ISAR成像。7、第七章提出了一种机载雷达对舰船目标的成像方法。在对回波序列进行运动补偿之后,利用广义二阶keystone变换同时结合孤立的散射点作WVD变换,估计多普勒参数并且构造相应的补偿函数来消除距离弯曲和距离走动,并最终实现舰船目标的高分辨成像。8、最后一章对全文工作进行了总结,并指出需要进一步研究的问题。更多还原

【Abstract】 Microwave imaging radar has an ability that can provide a high resolution radar image in all-weather, day/night and long range condition. Because the imaging radar can be competent for some missions that are difficult for some general surveillance systems, such as optics and infrared systems, it is being more and more widely used in military and civil fields. With the increased exploitation and utilization of ocean and space resource, the surveillance and safety protection of ocean and space resource will become more and more important. So, the detection and ISAR imaging of moving targets will play an important role in inspecting and identifying targets and have an important military and scientific research value.An emphasis will be put on the research of the detection and ISAR imaging of sea or space targets in the paper, which proposes a method for detecting wideband signal based on Crosscorrelation Function, builds the echo signal model of complexly moving targets, and proposes two methods based on the scale transform in time-frequency distribution plane (STTFD) and the time-chirp Clean (TC-DechirpClean) for targets ISAR imaging. Finally, a shown method is used for achieving ship imaging in airborne radar system. The main work of this dissertation is as follows:[1] Chapter 1 is the introduction. It reviews targets detection and ISAR development, then introduces the dissertation’s research background and main work.[2] Chapter 2 overviews conventional ISAR imaging principle and various technological methods used for ISAR imaging.[3] Chapter 3 studies a targets detection technology in wideband signal. The proposed parameter estimation algorithm is composed of the following steps: Firstly, do adjacent cross correlation between received signals to eliminate target’s translational motion effect, and reduce the range cell migration; then apply Keystone transform to eliminate the couple effect of range and cross range; next detect moving targets with coherent accumulation of correlation function after an scale transformation to each range cell data. The acceleration and the change of acceleration in range direction of targets can be estimated based on the peak position of the processed adjacent correlation function of echo signals. The velocity of target is estimated based on the peak position of the cross range accumulated signal after the high order phase term having been compensated, which is calculated based on the estimation acceleration.[4] Chapter 4 proposes an ISAR imaging algorithm based on linearly modulated frequency stepped (LMFS) signal. The method makes full use of the information of range envelope and azimuth Doppler to estimate the radial velocity and acceleration and makes compensation for motion parameters in the raw data. At the same time, this method was also applied to the imaging of ISAR in low signal to noise rate (SNR). At the end, the detailed imaging steps of LMFS ISAR are put forward and the simulated results with different SNR confirms this method.[5] Chapter 5 presents a new method based on a scale transform in time-frequency distribution plane (STTFD). After range envelop alignment, the data of each range cell is transmitted into the new ones in a time-frequency domain. A scale transform is applied to the time-frequency domain data to remove the coupling effect between signal time and the correlation function delay. After the above transforms, the bias distribution in the time-frequency plane of each scatterer signal result of the cross-range quadratic phase term is changed to a beeline distribution parallel to the time axis. Therefore, a high resolution ISAR image for a maneuvering moving target is obtained by the Fourier transform to the processed data. The proposed ISAR imaging algorithm is verified by simulation and raw radar data results.[6] Chapter 5 studies a new method based on the time-chirp distribution for imaging complexly moving targets. we first model the complex motion of ship target with cubic phase terms (parameterized on chirp rate and its change rate), then a new ISAR imaging method, referred to as TC-DechirpClean, is proposed, which estimates the chirp rate and the change rate of chirp rate of all scatters in the time-chirp distribution plane. Both numerical and experimental results are provided to demonstrate the performance of the proposed method.[7] Chapter 7 presents an imaging algorithm for achieving ship imaging in airborne radar system. An effective method based on an idea that uses motion compensation, general second-step keystone transform twice and at the same time, finds an isolated point and adopts WVD transform to estimate Doppler parameters for removing range walk and curvature. And what is more, the detailed deductions are given in this paper. Finally, the SAR image of moving ship is acquired in the azimuth frequency field. The simulation result of three moving point-targets and the imaging result of real ship data confirm this method.[8] Chapter 8 is the summary of the dissertation. It also discusses future research areas to be further studied.更多还原