Deep learning has been fully verified and accepted in the field of electromagnetic signal classification.However,in many specific scenarios,such as radio resource management for aircraft communications,labeled data are difficult to obtain,which makes the best deep learning methods at present seem almost powerless,because these methods need a large amount of labeled data for training.When the training dataset is small,it is highly possible to fall into overfitting,which causes performance degradation of the deep neural network.For few-shot electromagnetic signal classifi-cation,data augmentation is one of the most intuitive countermeasures.In this work,a generative adversarial network based on the data augmentation method is proposed to achieve better classifi-cation performance for electromagnetic signals.Based on the similarity principle,a screening mech-anism is established to obtain high-quality generated signals.Then,a data union augmentation algorithm is designed by introducing spatiotemporally flipped shapes of the signal.To verify the effectiveness of the proposed data augmentation algorithm,experiments are conducted on the RADIOML 2016.04C dataset and real-world ACARS dataset.The experimental results show that the proposed method significantly improves the performance of few-shot electromagnetic signal classification.

Huaji ZHOU;Jing BAI;Yiran WANG;Licheng JIAO;Shilian ZHENG;Weiguo SHEN;Jie XU;Xiaoniu YANG

School of Artificial Intelligence,Xidian University,Xi'an 710071,China;Science and Technology on Communication Information Security Control Laboratory,Jiaxing 314033,China

中国航空学报（英文版）

[1]Huaji ZHOU;Jing BAI;Yiran WANG;Licheng JIAO;Shilian ZHENG;Weiguo SHEN;Jie XU;Xiaoniu YANG-.Few-shot electromagnetic signal classification:A data union augmentation method)[J].中国航空学报（英文版）,2022(09):49-57

powerless,RADIOML

Few,shot,electromagnetic,classification,union,augmentation,Deep,learning,has,been,fully,verified,accepted,field,However,many,specific,scenarios,such,radio,resource,management,aircraft,communications,labeled,are,difficult,obtain,which,makes,best,deep,methods,present,seem,almost,because,these,need,large,amount,training,When,dataset,small,highly,possible,fall,into,overfitting,causes,performance,degradation,neural,network,For,few,one,intuitive,countermeasures,In,this,generative,adversarial,proposed,achieve,better,signals,Based,similarity,principle,screening,mech,anism,established,quality,generated,Then,algorithm,designed,by,introducing,spatiotemporally,flipped,shapes,To,verify,effectiveness,experiments,conducted,04C,real,world,ACARS,experimental,results,show,that,significantly,improves

0.558612