When Internet of Things(IoT)nodes ac-cess the network through wireless channels,the net-work is vulnerable to spoofing attacks and the Sybil attack.However,the connection of massive devices in IoT makes it difficult to manage and distribute keys,thus limiting the application of traditional high-level authentication schemes.Compared with the high-level authentication scheme,the physical layer authentica-tion scheme realizes the lightweight authentication of users by comparing the wireless channel characteris-tics of adjacent packets.However,traditional physical layer authentication schemes still adopt the one-to-one authentication method,which will consume numer-ous network resources in the face of large-scale IoT node access authentication.In order to realize the se-cure access authentication of IoT nodes and regional intrusion detection with low resource consumption,we propose a physical layer authentication mechanism based on convolution neural network(CNN),which uses the deep characteristics of channel state informa-tion(CSI)to identify sending nodes in different loca-tions.Specifically,we obtain the instantaneous CSI data of IoT sending nodes at different locations in the pre-set area,and then feed them into CNN for training to procure a model for IoT node authentication.With its powerful ability of data analysis and feature extrac-tion,CNN can extract deep Spatio-temporal environ-ment features of CSI data and bind them with node identities.Accordingly,an authentication mechanism which can distinguish the identity types of IoT nodes located in different positions is established to authen-ticate the identity of unknown nodes when they break into the pre-set area.Experimental results show that this authentication mechanism can still achieve 94.7％authentication accuracy in the case of a low signal-to-noise ratio(SNR)of 0 dB,which means a signifi-cant improvement in authentication accuracy and ro-bustness.

Xinglu Li;Kaizhi Huang;Shaoyu Wang;Xiaoming Xu

PLA Strategic Support Force Information Engineering University,Zhengzhou 450001,China

中国通信（英文版）

[1]Xinglu Li;Kaizhi Huang;Shaoyu Wang;Xiaoming Xu-.A Physical Layer Authentication Mechanism for IoT Devices)[J].中国通信（英文版）,2022(05):129-140

authentica,authen,ticate

Physical,Layer,Authentication,Mechanism,IoT,Devices,When,Internet,Things,nodes,network,through,wireless,channels,vulnerable,spoofing,attacks,Sybil,However,connection,massive,devices,makes,difficult,manage,distribute,keys,thus,limiting,application,traditional,high,level,authentication,schemes,Compared,physical,layer,realizes,lightweight,users,by,comparing,adjacent,packets,still,adopt,one,method,which,will,consume,numer,resources,face,large,scale,access,order,regional,intrusion,detection,low,consumption,propose,mechanism,convolution,neural,uses,deep,characteristics,state,informa,CSI,identify,sending,different,Specifically,obtain,instantaneous,data,locations,pre,set,area,feed,them,into,training,procure,model,With,its,powerful,ability,analysis,extract,Spatio,temporal,environ,features,bind,identities,Accordingly,distinguish,identity,types,located,positions,established,unknown,when,they,break,Experimental,results,show,that,this,achieve,accuracy,case,signal,noise,ratio,SNR,dB,means,signifi,cant,improvement,bustness

0.495971