基于随机卷积原型网络的雷达辐射源个体识别  

Specific Radar Emitter Identification Based on Random Convolutional Prototypical Networks

作　　者：孙佳杰 崔良中[1] 牛雅萌 陆剑雄 漆健驰 SUN Jiajie;CUI Liangzhong;NIU Yameng;LU Jianxiong;QI Jianchi(School of Electronic Engineering,Naval University of Engineering,Wuhan 430033)

机构地区：[1]海军工程大学电子工程学院,武汉430033

出　　处：《舰船电子工程》2024年第12期82-86,共5页Ship Electronic Engineering

摘　　要：针对传统雷达辐射源个体识别方法准确率低,时效性差,泛化性不强的问题,论文提出基于随机卷积原型网络的雷达辐射源个体识别方法。该方法通过利用大量随机卷积核对雷达辐射源信号进行高效地特征提取变换,建立原型网络模型,将变换特征输入到原型网络中进行原型学习,令原型网络为每一类雷达辐射源个体学习一个原型。通过原型网络变换的辐射源特征到原型距离作为分类依据,完成雷达辐射源个体识别。对6类雷达辐射源个体采集进行实验,结果表明:该方法有效解决了雷达辐射源个体识别问题。This paper proposes a specific radar emitter identification method based on random convolutional prototype network to solve the problems of low accuracy,poor timeliness and poor generalization of traditional specific radar emitter identification methods.The method uses a large number of random convolution kernels to extract and transform features of radar emitter signals ef⁃ficiently,establishes a prototype network model,inputs the transformed features into the prototype network for prototype learning,and makes the prototype network learn a prototype for each type of radar emitter signal.The distance is used as the classification ba⁃sis.Specific radar emitter identification is completed by the distance between the emitter signal features and the prototype trans⁃formed by the prototype network.Experiments are carried out on six types of radar emitters.The results show that the proposed meth⁃od effectively solves the problem of specific radar emitter identification.

关 键 词：雷达辐射源个体识别 特征提取 随机卷积核 原型学习 

分 类 号：TN957.51[电子电信—信号与信息处理] TP183[电子电信—信息与通信工程]