Cramer-Rao bound and signal-to-noise ratio gain in distributed coherent aperture radar

This paper studies the estimation performance of the coherent processing parameter （CPP）, including time delay differences and phase synchronization errors among different apertures of the distributed coherent aperture radar （DCAR）. Firstly, three architectures of signal processing in the DCAR are introduced. Secondly, the closed-form Cramer-Rao bound （CRB） of the CPP estimation is derived and compared. Then, the closed-form CRB is verified by numerical simulations. Finally, when the next generation radar works in a fully coherent mode, the closed-form signal-to-noise ratio （SNR） gain of the three architectures is presented.     

Knowledge-based detection method for SAR targets

When the classical constant false-alarm rate （CFAR） combined with fuzzy C-means （FCM） algorithm is applied to target detection in synthetic aperture radar （SAR） images with complex background, CFAR requires block-by-block estimation of clutter models and FCM clustering converges to local optimum. To address these problems, this paper pro-poses a new detection algorithm： knowledge-based combined with improved genetic algorithm-fuzzy C-means （GA-FCM） algorithm. Firstly, the algorithm takes target region＇s maximum and average intensity, area, length of long axis and long-to-short axis ratio of the external ellipse as factors which influence the target appearing probabil- ity. The knowledge-based detection algorithm can produce preprocess results without the need of estimation of clutter models as CFAR does. Afterward the GA-FCM algorithm is improved to cluster pre-process results. It has advantages of incorporating global optimizing ability of GA and local optimizing ability of FCM, which will further eliminate false alarms and get better results. The effectiveness of the proposed technique is experimentally validated with real SAR images.     

Angular extent effect of micromotion target in SAR image by polar format algorithm

Target micromotion not only plays an important role in target recognition but also leads to esoteric characteristics in synthetic aperture radar （SAR） imaging. This paper finds out an interesting phenomenon, i.e. the angular extent effect, in micro-motion target images formulated by the polar format algorithm. A micromotion target takes on multiple pairs of paired echoes （PEs） around the true point, and each PE extends for an angle which is exactly equal to the angular extent of the synthetic aperture, regardless of the micromotion frequency. The effect is derived and interpreted by using the characteristics of Bessel functions. Then it is demonstrated by simulation experiments of a target with different micromotion frequencies. The revelation and interpretation of the effect is highly beneficial to micromotion-target SAR image understanding as wel as target recognition.     

Grating lobes suppression method for stepped frequency GB-SAR system

A grating lobes suppression method for chirp-subpulse stepped frequency（CSSF） signals is proposed, which is applied to deformation monitoring using the ground based synthetic aperture radar（GB-SAR） system. This method is based on accurate estimation and correction of the phase and amplitude error along two dimensions（range and azimuth）, i.e., the error estimation inside the subpulse（in-subpulse error） and across the stepped frequency subpulses（cross-subpulse error） of transmitted CSSF signals. Validated both with simulated data and experimental data recorded in the deformation monitoring campaign, it can be seen that the method as well as the relative conclusions can be fully and effectively applied to most of the stepped frequency systems.     

数学形态学在SAR图像目标识别中的应用

提出了一种能够从复杂背景的SAR图像中识别出圆状目标的方法。该方法首先对SAR图像进行双参数恒虚警(B-CFAR)预处理,排除明显的非目标区,增强下一步对目标的识别效果,然后着重分析了上下文相关加权的形态模板生成算法,在此基础上采用不同的数学形态学变换滤波方法对预处理后的SAR图像候选目标区进行识别,并与相关文献比较,对圆状目标得到了良好的识别效果。     

SVM和HMM相结合的合成孔径雷达图像目标识别

提出了一种支持向量机和隐马尔可夫模型相结合的合成孔径雷达图像目标识别方法。该方法用小波分解和主成分分析提取图像特征,生成特征向量。利用图像在方位角上的关系由特征向量生成图像的特征序列以及隐马尔可夫模型的训练序列。用支持向量机进行目标预识别,确定目标最有可能所属的两个类别,用隐马尔可夫模型在这两个类别中确定目标最终所属类别,完成目标识别。使用MSTAR数据库中的图像数据对该方法进行验证和分析,结果表明,该方法可以明显提高目标的正确识别率,是一种有效的合成孔径雷达图像目标识别方法。     

一种改进的二端口杂波抑制干涉仪

分析了二端口杂波抑制干涉仪的工作原理 ,讨论了系统中的误差及噪声对主杂波抑制的影响。在此基础上 ,对二端口杂波抑制干涉仪方案中最关键的对消因作做了改进。分析和仿真结果表明 :经改进的二端口杂波抑制 ,干涉仪对误差及噪声的敏感程度显著下降 ,能够稳健地抑制主杂波 ,进而使检测淹没在主瓣杂波中的慢动军事目标成为可能。     

基于频域稀疏压缩感知的雷达目标三维散射中心反演方法

基于平面阵列的微波天线结构可以获取多视角下的目标散射中心三维分布。针对平面阵列稀疏分布导致的目标成像及散射中心反演精度较差的问题, 设计了一种基于组合巴克码的稀疏孔径分布方式。在此基础上, 利用稀疏孔径回波和频域主成分分析得到参考复数信号。利用该参考复数信号对原始回波进行干涉处理, 获得回波频谱的稀疏表征方式。在频域建立基于压缩感知的目标散射中心三维分布模型并进行优化求解, 得到重构后的目标三维频谱, 并逆变换至空间域, 可实现目标散射中心幅度及三维位置重建。暗室试验数据处理结果表明, 所提方法在X波段和稀疏采样率为50%的条件下, 目标散射中心幅度及三维位置反演精度均优于90%。     

天然气汽车改装中CNG喷射器喷嘴孔径的确定

在压缩天然气汽车(CNGV)的改装过程中,CNG喷射器喷嘴孔径这个工艺参数,极为关键却又难以把握。为了提高安装作业的效率及精度,探究出一种能辅助安装者更快更准地确定所需孔径的方法,提出了以模糊模式识别原理的算法,根据待改装车辆的发动机功率、行驶总里程和当前火花塞使用中的行驶里程这3个指标,对喷嘴孔径的尺寸进行识别,以实现在CNG系统安装前对孔径进行确定。在MATLAB中对该算法进行验证时,发现采用随机性更好的分组抽样且对算法的数据运算过程进一步匹配修正后,孔径的识别精准性显著提高。经过与此前的实际安装记录数据的核对,最终证实了该算法的可行性及准确性。     

基于改进的CNN和数据增强的SAR目标识别

针对合成孔径雷达(synthetic aperture radar, SAR)图像目标识别问题, 提出了基于改进的卷积神经网络和数据增强的SAR目标识别方法。首先在训练阶段引入Dropout, 随机删除部分神经元, 增强网络的泛化能力。其次, 在网络中引入L2正则化, 简化模型的同时降低结构风险, 并且能有效地抑制过拟合。然后, 采用Adam优化网络, 提高模型的收敛效率。最后, 采用优选的数据增强方法, 扩充SAR目标数据集, 为网络训练提供更为充足的样本, 进一步提高识别的准确率和模型的泛化性。在运动和静止目标获取与识别(moving and stationary target acquisition and recognition, MSTAR)数据集上进行了实验, 结果表明设计的卷积神经网络识别准确率高, 且具有更好的泛化性。