基于点阵图模型的多功能雷达波形单元提取方法  

Multi-Function Radar Waveform Unit Extraction Method Based on Dot Plot Model

作　　者：付雨欣 蒋能 黄洁[1] 张红敏[1] 王建涛 FU Yuxin;JIANG Neng;HUANG Jie;ZHANG Hongmin;WANG Jiantao(Information Engineering University,Zhengzhou 450001,China)

机构地区：[1]信息工程大学,河南郑州450001

出　　处：《信息工程大学学报》2023年第3期265-271,共7页Journal of Information Engineering University

基　　金：国家自然科学基金面上资助项目(62071490);河南省自然科学优秀青年基金资助项目(212300410095)。

摘　　要：针对先验知识缺失情况下非协作多功能雷达(Multi-Function Radar,MFR)波形单元提取问题,提出一种基于点阵图模型的波形单元提取方法。首先,在分析MFR脉冲列特点的基础上,构建了MFR脉冲列点阵图模型,直观地显示出波形单元起始脉冲位置;其次,采用孔洞填充、闭运算等图像预处理方法消除噪声对点阵图的影响;最后,利用Harris角点检测算子识别波形单元起始点。仿真实验表明,该方法相比于DA-DBSCAN波形单元提取算法,检测概率和F值有明显提升,虚警概率显著下降,在脉冲丢失和参数测量误差较大时仍具有较强的鲁棒性。To address the problem of non-cooperative multi-function radar(MFR)waveform unit ex-traction in the absence of prior knowledge,a waveform unit extraction method based on dot plot model is proposed.Firstly,this method constructs a dot plot model of MFR pulse train,which can display the initial pulse position of waveform unit visually through analyzing the characteristics of MFR pulse train.Then,image preprocessing methods such as hole-filling and closing operation are used to remove the influence of noise on this model.Finally,Harris corner detection is used to iden-tify the initial point of waveform unit.The simulation results show that compared with DA-DBSCAN waveform unit extraction algorithm,the detection probability and F score of the proposed method are significantly improved,and the false alarm probability is significantly reduced.Meanwhile,it still has strong robustness against pulse loss and large parameter measurement error.

关 键 词：多功能雷达 波形单元 点阵图 角点检测 图像处理 

分 类 号：TN971.1[电子电信—信号与信息处理]