雷达高分辨距离像自适应角域划分方法

发布时间：2018-06-21 00:21

本文选题：高分辨距离像 + 功率谱互相关系数　；参考：《系统工程与电子技术》2014年11期

【摘要】：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。
[Abstract]:Radar high resolution range profile radar (HRRP) has the characteristic of attitude sensitivity. One effective method is to deal with a group of HRRP-samples in the range of angle range in which no scattering points cross the range of moving units. A sequential discriminant adaptive framing algorithm based on HRRP echo power spectrum is proposed. The boundary of data partition is determined by iterative search of global HRRP samples and the number of angular domains is divided adaptively. According to the theory of scattering point model and considering the law of power spectrum cross-relation, the proposed method can effectively restrain the sample of HRRP with "range profile fluctuating intensely". Compared with the adaptive Gao Si classifier, the proposed method can effectively partition the angle domain of the target even if the sample data is small. At the same time, compared with the traditional methods, the proposed method can reduce the recognition computation and processing time of the recognition algorithm effectively. The simulation of three types of ship targets and the analysis of field data of three types of civil ships show that the proposed method is effective.
【作者单位】：海军航空工程学院电子信息工程系;海军装备部;
【基金】：上海市科学技术委员会(13ZR1440600)资助课题
【分类号】：TN957.52

【参考文献】