运动平台前视雷达超分辨成像理论与方法

发布时间：2018-06-28 21:36

本文选题：运动平台 + 前视扫描雷达　；参考：《电子科技大学》2016年博士论文

【摘要】：运动平台载雷达实波束扫描图像,受天线孔径限制,方位角分辨率难以满足精确末制导、地形回避和自主着陆等应用的要求。而现有的前视单脉冲锐化技术虽然改善了图像显示效果,但不能提高图像分辨能力。因此,开展运动平台雷达前视成像方法研究,提出改善角分辨率的超分辨方法,突破实波束分辨率的制约,具有重要的应用价值。本文围绕运动平台载雷达前视超分辨成像问题,开展了回波建模、超分辨算法和飞行实验验证等研究工作,主要内容如下:1.导出了扫描雷达的方位类卷积信号模型,为采用卷积反演方法实现前视超分辨提供了理论支撑。还研究了超分辨性能的信噪比约束条件,为工程应用中确定雷达工作参数提供了理论依据。2.提出了稀疏目标超分辨成像算法,可有效抑制卷积反演过程中的噪声和杂波放大效应,避免出现虚假目标。显著改善了噪声和杂波背景中孤立多目标的分辨能力。3.提出了面目标超分辨成像算法,以广义高斯分布作为目标先验信息,并通过调节其分散度参数确定目标函数,可实现噪声和杂波背景下的面目标超分辨成像,同时改善面目标图像纹理信息。4.提出了等效阵列前视超分辨成像算法,将前视扫描转化为阵列模型,并以最小二乘和最小均方误差准则进行阵列超分辨。可以在少量方位扫描次数条件下,实现孤立目标个数和位置的准确估计。上述模型和超分辨成像算法,已通过仿真或实测数据进行了验证,结果表明,论文提出的超分辨成像算法,可实现实际运动平台载雷达前视超分辨成像。
[Abstract]:The real beam scanning images of radar on moving platform are limited by antenna aperture and the azimuth resolution is difficult to meet the requirements of precise terminal guidance terrain avoidance and autonomous landing. Although the existing forward looking monopulse sharpening technology can improve the image display effect, it can not improve the image resolution. Therefore, it is of great value to research forward imaging method of moving platform radar, to improve the angle resolution and to break through the constraints of real beam resolution. In this paper, the echo modeling, super-resolution algorithm and flight experiment verification are carried out around the problem of super-resolution imaging of moving platform-borne radar. The main contents are as follows: 1. The azimuth type convolution signal model of scanning radar is derived, which provides theoretical support for using convolution inversion method to realize forward looking super resolution. The signal-to-noise ratio (SNR) constraint condition of superresolution performance is also studied, which provides a theoretical basis for determining the operational parameters of radar in engineering applications. A super-resolution imaging algorithm for sparse targets is proposed, which can effectively suppress the noise and clutter amplification in convolution inversion and avoid the appearance of false targets. The resolution ability of isolated multi-target in noise and clutter background is improved significantly. In this paper, a super-resolution imaging algorithm for surface targets is proposed. The generalized Gao Si distribution is used as the prior information of the targets, and the target function is determined by adjusting the dispersion parameters. The super-resolution imaging of surface targets under noise and clutter background can be realized. At the same time, improve the face target image texture information. 4. In this paper, an equivalent array forward super-resolution imaging algorithm is proposed. The forward scan is transformed into an array model, and the array superresolution is carried out by using the least square method and the least mean square error criterion. The accurate estimation of the number and position of isolated targets can be realized under the condition of a few azimuth scanning times. The above model and super-resolution imaging algorithm have been verified by simulation or measured data. The results show that the super-resolution imaging algorithm proposed in this paper can realize the super-resolution imaging of radar on the actual moving platform.
【学位授予单位】：电子科技大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN957.52

【参考文献】