【作者】 毛勇；

【导师】 阮成礼；

【作者基本信息】 电子科技大学 ， 电磁场与微波技术， 2007， 博士

【摘要】 成像雷达作为一种先进的探测工具,能全天时、全天候、远距离提供高分辨率的雷达图像,在军用和民用的众多领域具有广泛的应用前景。为获得高分辨雷达图像,回波信号间必须保持精确的相位关系,但是雷达与散射点间的相对运动会引入一定的初相误差,破坏雷达信号间的相位相干性,从而影响所得图像质量。因此为保持雷达信号间的相位相干性,必须进行精确的运动补偿。本文在实测雷达数据处理中发现,当群体目标高速机动飞行时,现有的运动补偿算法和成像算法的应用都受到很大的限制。为了满足高速机动多目标成像的需要,本文进行了相应的研究,提出了机动多目标相对运动的运动补偿算法以及目标自旋运动的运动补偿算法,并通过实验与仿真数据的处理予以验证。本文的主要贡献如下:提出了一种新的相对运动补偿算法。常规运动补偿算法,包括相关法及最小熵方法,均是基于目标为刚体进行的,当多目标间存在着相对运动时,直接利用常规补偿方法通常得不到满意的结果。针对这一问题,本章建立了目标相对运动的模型,分析相对运动对成像的影响,提出了一种新的相对运动补偿算法。该算法对存在相对运动目标在距离上利用Keystone变换进行分离,提取相对运动目标的回波分别利用auto-cleans算法进行运动补偿。模拟和实测数据的成像结果说明该算法是有效的。提出了一种新的自旋运动补偿算法。建立了自旋运动目标的运动模型,研究自旋运动对ISAR成像的影响,推导出自旋运动产生的多普勒频率的表达式,采用基于“CLEAN”多分量线性调频信号的分解以及AJTF技术对回波相位进行参数估计,去掉旋转产生的相位高次项的影响,消除自旋对ISAR成像的影响,仿真及实验结果表明该补偿方法效果明显。利用改进的时频分布级数法提高图像分辨能力。时频分布级数法(TFDS)在保持高时频分辨率的前提下,有效抑制了Wigner分布的交叉项,但算法的实现速度已经成为制约TFDS方法在实际中推广的关键问题。本章一方面采用实数形式代替复数形式的Gabor展开,另一方面去掉对时频原子的冗余计算,大大降低了运算量。通过对多分量非平稳信号的仿真表明,对TFDS方法的改进是有效的,信号处理时间明显缩短,可用于对目标回波时频特征的快速提取。研制了数据处理软件,在工程上实现高速群体运动目标ISAR二维成像。利用现有窄带雷达系统的数据接口,设计了一种用于改造数据采集系统的I/Q解调器,在现有窄带单脉冲雷达给出的目标回波和目标运动轨迹数据的基础上,利用前几章研究的算法,编制数据处理软件,实现高速运动多目标的距离压缩、运动补偿、横向成像处理、RCS定标。本文在以上工作的基础上,完成了对高速运动群体目标目标特性(RCS)的动态测量,并对结果动态显示。实验表明本文的ISAR成像结果与光学成像一致。更多还原

【Abstract】 Imaging radar, which can provide an all-weather day/night capability to generate a high-resolution representation of radar reflectivity of targets over a long range, has received intensive attention because of its potential applications in both civilian and military applications. In order to obtain high-resolution radar images, it is necessary to maintain the precise phase relationship between return signals. This requires precise motion compensation, because the unplanned perturbations of the relative motion between the radar and the scatters causes phase errors, which may destroy the phase coherent of return signals and then blur the image.The study on the raw radar data shows that the traditional algorithms of motion compensation and imaging to ISAR available cannot suit to high-speed maneuvering targets. In this paper some algorithms of motion compensation and imaging to high-speed maneuvering targets are proposed and verified by raw and simulation data.The main contributions of the dissertation are as following:A new algorithm for relative motion compensate is proposed. Because conventional motion compensation methods assume that targets’ positions are almost fixed during the imaging time, which requires the targets be rigid. It is not easy for us to obtain good result for several relative motion targets using these methods. In this chapter the model of relative motion targets is established, and a new algorithm for relative motion compensation based on Keystone transformation and auto-clean algorithm is proposed. The imaging results based on simulation and real data validate the new relative motion compensation method.A new algorithm for spinning motion compensation is presented. ISAR imaging of a self-spinning motion target is described, and the effect of self-spinning motion for ISAR image is analyzed, the Doppler-frequency expression produced by self-spinning motion is deducted. Using multi-component linear frequency-modulated signals decomposition method based on "CLEAN" in frequency domain and AJTF to estimate parameter of echo’s phase, we can remove the influence of high step phase produced by spin and attain clarity ISAR images of self-spinning targets.Using the improved time-frequency distribution series (TFDS) to improve Imaging resolution. The TFDS has been tested effectively in the balance of the cross-term suppression and the time-frequency resolution. We replace the complex form discrete Gabor transform with the real one to reduce computing. On the other hand, we get ride of the redundant computing about the time-frequency atom. The simulation of the multi-component non-station signal shows that the improvement of the TFDS method accelerates the computing.Exploiting the data processing software to realize RCS dynamic measure of high-speed maneuvering targets. In this chapter, using the data port of the narrowband radar system available, we have designed a sort of I/Q demodulator to alter the data sampling system. Based on received signals of narrow band single pulse radar and its orbit data, using forenamed algorithms, we exploit the data processing software, which realizes range compression and relative motion compensation and cross-rang imaging of high-speed maneuvering targets.Based on forenamed works of the dissertation, the whole process can display imaging results and RCS dynamic measure of high-speed maneuvering targets is achieved. ISAR imaging results of experiments are consistent with optical radar images.更多还原