基于霍夫变换的旋转长基线干涉仪雷达信号分选算法  

Radar signal deinterleaving algorithm for rotating long baseline interferom-eter based on Hough transform

作　　者：李江浩 季华益[1] 罗佳奕 周钧 余仲阳 李煊鹏 Li Jianghao;Ji Huayi;Luo Jiayi;Zhou Jun;Yu Zhongyang;Li Xuanpeng(No.8511 Research Institute of CASIC,Nanjing 210007,Jiangsu,China;School of Instrument Science and Engineering,Southeast University,Nanjing 211189,Jiangsu,China)

机构地区：[1]中国航天科工集团8511研究所,江苏南京210007 [2]东南大学仪器科学与工程学院,江苏南京211189

出　　处：《航天电子对抗》2022年第2期21-27,共7页Aerospace Electronic Warfare

摘　　要：信号分选是雷达信号处理的基础与前提,传统方法在雷达数量增加、模式多变的情况下,分选性能显著下降。通过分析旋转长基线干涉仪(RLBI)体制中辐射源脉冲相位参数时变的特性,提出了一种基于霍夫变换的信号分选方法。该方法首先利用频率参数聚类进行初次分选,再利用霍夫变换提取不同来波方向上脉冲的相位时变特性直线,最后利用相位周期延拓进行脉冲序列提取。仿真实验表明了算法的有效性。相比传统方法,该方法能够利用相位参数直接分选不同来波方向上的脉冲,能在一定程度上解决旋转干涉仪信号分选中因缺乏方位信息而造成的分选效果下降问题。Radar signal deinterleaving is the basis and premise for radar signal processing.Due to increasing amount and flexible working modes of radar,the performance of conventional signal deinterleaving algorithm degrades seriously.Through the analysis of the relationship between pulse phase and Time of Arrival(TOA)in Rotating Long Baseline Interferometer(RLBI)while the baseline rotating,a new deinterleaving method for rotating long baseline interferometer is proposed.Firstly it classifies pulses based on frequency parameter by grid clustering method.Then it extracts line feature between phase and TOA of pulses from different directions by Hough transform.Lastly it extracts pulse sequence of each radar by periodic extension of phase parameter.The effectiveness of this algorithm is validated by simulated experiments.Comparing to conventional signal deinter⁃leaving algorithms in rotating RLBI,this method is able to classify pulses in different directions of arrival directly before direction finding of each radar,which can solve the problem of the degrading deinterleaving performance caused by the lack of direction information of a single pulse.

关 键 词：雷达信号分选 旋转长基线干涉仪 网格聚类算法 霍夫变换 

分 类 号：TN971[电子电信—信号与信息处理]