空天预警研究学报2025，Vol.39Issue(6)：409-414,6.DOI:10.3969/j.issn.2097-180X.2025.06.004
基于深度频率动态卷积神经网络的干扰样式识别研究
Research on interference pattern recognition based on deep frequency dynamic convolutional neural network
周泽涛 1杜庆磊2
作者信息
- 1. 空军预警学院,武汉 430019||94969部队,上海 200000
- 2. 空军预警学院,武汉 430019
- 折叠
摘要
Abstract
With the complication and diversification of interference patterns in the modern electronic counter-measure environment,traditional interference identification methods have faced bottlenecks in practical applica-tions.In order to address the interference identification in complex environments,this paper proposes a deep neu-ral network based on frequency dynamic convolution for identifying seven typical interference patterns.First,con-tinuous wavelet transform is performed on the interference signal to extract two-dimensional time-frequency fea-ture maps.Then,the deep frequency dynamic convolutional neural network(D-FDConvNet)is used for feature ex-traction and interference pattern recognition in the feature maps.Simulation results demonstrate that the proposed method maintains a high recognition accuracy rate even at low jamming-to-noise ratio(JNR),and that its effect is superior to the traditional convolutional neural network,which validates its effectiveness and superiority in the in-terference recognition task.关键词
干扰识别/频率动态卷积/卷积神经网络/小波变换/特征提取Key words
interference recognition/frequency dynamic convolution/convolutional neural network(CNN)/wavelet transform/feature extraction分类
信息技术与安全科学引用本文复制引用
周泽涛,杜庆磊..基于深度频率动态卷积神经网络的干扰样式识别研究[J].空天预警研究学报,2025,39(6):409-414,6.
