雷达高分辨距离像自适应角域划分方法

雷达高分辨距离像(radar high resolution range profile， HRRP)具有姿态敏感性的特点，一种有效的方法是分别对一组不发生散射点越距离单元走动角域范围内的HRRP样本进行处理。提出基于HRRP回波功率谱的顺序判别自适应分帧算法，通过对全局HRRP样本进行迭代搜索来确定数据划分边界，并自适应划分角域个数。所提方法依据散射点模型理论，考虑功率谱互相关系数变化规律，有效抑制了HRRP存在“距离像起伏剧烈”的样本。与自适应高斯分类器划分角域方法相比，所提方法在样本数据较小的情况下，仍可以有效地对目标进行角域划分。同时，与传统的利用互相关系数分帧方法和等间隔角域划分方法相比，所提方法可以有效降低识别运算量，减少识别算法处理时间，通过对3类舰船目标的仿真和3类民用船只的外场实测数据分析,表明所提方法是有效的。

Adaptive angular-sector segmentation method for radar HRRP

Radar high resolution range profile (HRRP) is very sensitive to target gesture variation. An effective method is to deal with the HRRPs within an angular sector boundary in which a set of scattering points will not migrate through resolution cells. A mathematical model based on the power spectrum cross correlation coefficient method is proposed,which determines the target aspect sector boundary based on the global HRRPs with the iteration algorithm,and the target aspect sector number can be determined simultaneously. Based on the scattering points model theory, the proposed method takes into consideration the change rule of the power spectrum cross correlation coefficient to effectively suppress HRRPs’“violent fluctuation” samples. Compared with the adaptive Gaussian classifier angular sector segmentation method, the proposed method can effectively divide targets’ angular sector with less samples. Compared with the traditional frame segmentation method based on the cross correlation coefficient and the traditional uniform frame segmentation method, the proposed method can effectively reduce the recognition time and computation complexity. Experimental results based on three types of warships and three kinds of civilian vessels field measured data show the efficiency of the proposed method.