摘要
为了提升雷达辐射源个体识别方法的识别精度,提出利用脉冲信号的载波频率特征进行辐射源识别。对于实测获得的脉冲信号数据,通过FFT实现时频域转换,并完成对15种频域统计特征的提取,进而通过PCA算法筛选出其中10种最能代表个体间差异性的指纹特征,最后将提取出的特征作为输入,通过自主构建的三种分类器实现个体识别。实验结果表明频域特征能较好地应用于辐射源分类识别。
In order to improve the recognition accuracy of the radar radiation source identification method,the carrier frequency characteristics of the pulse signal are proposed to identify the radiation source.For the measured pulse signal data,the time-frequency domain conversion is realized through FFT,and 15 kinds of frequency domain statistical features are extracted,and then 10 kinds of fingerprint features that can best represent the difference between individuals are screened by PCA algorithm.Finally,the extracted features are used as input to realize individual recognition through three kinds of independently constructed classifiers.The experimental results show that the frequency domain features can be applied to the classification and identification of radiation sources.
作者
兰博森
李胜勇
陈剑
于凯翔
李洪科
LAN Bosen;LI Shengyong;CHEN Jian;YU Kaixiang;LI Hongke(School of Electrical Engineering,Naval University of Engineering,Wuhan 430033)
出处
《舰船电子工程》
2024年第9期74-77,123,共5页
Ship Electronic Engineering
关键词
雷达辐射源
个体识别
载波频率
特征提取
radar radiation source
individual recognition
carrier frequency
feature extraction
