基于改进矩阵束法的雷达目标回波极点提取方法  

Pole extraction of radar target echo based on improved matrix pencil algorithm

作　　者：宋歌 贾鑫 张勇顺 毛鑫峰 SONG Ge;JIA Xin;ZHANG Yong-shun;MAO Xin-feng(Equipment College,Beijing 101416,China)

机构地区：[1]装备学院,北京101416

出　　处：《电子设计工程》2018年第13期86-89,94,共5页Electronic Design Engineering

摘　　要：目标极点因具有方位不变性而成为雷达目标识别的重要研究对象。为进一步改善矩阵束法在低信噪比条件下的极点提取精度,提出了一种基于改进矩阵束法和FEKO软件仿真的雷达目标回波极点提取方法,将奇异值分解法、互相关处理法与传统的矩阵束法相结合,仿真得到了理想目标和导弹目标的回波极点,并进行了误差分析。结果表明,针对信噪比大于10 dB的回波信号,这种方法的极点提取准确度可以达到95%以上,对模拟导弹目标回波的极点提取结果可以验证极点的方位不变性。Target pole becomes an important research object in radar target identification because of its aspectual Invariance. In order to improve the pole extraction accuracy under low signal to noise ratio condition,a pole extraction method of radar target echo based on improved matrix pencil algorithm and FEKO simulation is proposed. The traditional matrix algorithm has been improved by using and singular value decomposition and cross-correlation-based processing methods. Through simulation,poles of ideal and missile targets were acquired and errors were analyzed. The results show that the pole extraction accuracy can be as high as 95% aim at the echo signal with signal-to noise ratio larger than 10 dB. The aspectual Invariance can be validated by the Pole extraction results of missile target echo.

关 键 词：极点 矩阵束法 奇异值分解 互相关处理 方位不变性 

分 类 号：TN957.52[电子电信—信号与信息处理]