基于深度残差网络的脉内调制信号识别研究  

Research on intra-pulse modulation signal recognition based on deep residual network

作　　者：邵广庆 陈歆伟 包军 周钧 许海波 晏杨 Shao Guangqing;Chen Xinwei;Bao Jun;Zhou Jun;Xu Haibo;Yan Yang(No.8511 Research Institute of CASIC,Nanjing 210007,Jiangsu,China)

机构地区：[1]中国航天科工集团8511研究所,江苏南京210007

出　　处：《航天电子对抗》2022年第3期39-44,共6页Aerospace Electronic Warfare

摘　　要：针对低信噪比下,传统方法对复杂雷达脉内调制信号类型识别率低的问题,提出了一种基于深度残差网络的复杂雷达脉内调制信号类型自动识别方法。该识别方法的基本原理是通过短时傅里叶变换(STFT)获取复杂脉内调制信号的时间与频率的变换关系,然后设计恰当的深度残差网络对时频图进行特征提取并完成多种复杂脉内调制信号类型识别。此外,为了增强识别方法的泛化能力,引入了标签平滑和L2正则化操作。仿真结果表明,该方法在低信噪(2 dB)下,对复杂雷达脉内调制信号类型的识别精度达到95%以上。In view of low recognition rate of complex radar intra-pulse modulation signal type by traditional methods under low signal-to-noise ratio(SNR),an automatic recognition method of complex radar intra-pulse modulation signal type based on deep residual network is proposed.The basic principle of the recognition method is to obtain the transformation relationship between the time and frequency of complex radar intra-pulse modulation signal through short-time Fourier transform(STFT),and then an appropriate deep residual network to extract the features of the time-frequency map is designed and a variety of complex intra-pulse modulation signal type recognition is completed.In addition,in order to improve the generalization ability of proposed method,label smoothing and L2 regularization are introduced.The simulation results show that the proposed method has a recognition accuracy of more than 95% for complex radar intra-pulse modulation signal types under low SNR(2 dB).

关 键 词：脉内调制 低信噪比 深度残差网络 自动识别 

分 类 号：TN971.1[电子电信—信号与信息处理] TN974[电子电信—信息与通信工程]