Research On Access Authentication Technology Based On Channel Information In Wireless Network

Because of its proximity access characteristics,edge computing can achieve faster network service response and meet the needs of real-time business control,security and privacy protection in multiple industries.In the process of massive heterogeneous terminals accessing the network through edge devices,achieving fast and reliable authentication is extremely challenging.The physical layer authentication technology is authenticated by the channel characteristics,and has a lightweight feature that can be well adapted to the authentication scenario of massive heterogeneous terminal access.The asymmetric resources provided by edge computing are especially suitable for access authentication of physical layer technologies.In this paper,the wireless communication access authentication technology under the edge calculation has completed the following aspects:In order to solve the problem that the threshold is difficult to find,a physical layer channel authentication scheme based on AdaBoost is proposed.Combining the physical layer authentication and the AdaBoost algorithm,a one-dimensional feature-based authentication model is established to realize the dynamic determination of the threshold.In the case of adopting one-dimensional features,compared with the manual traversal to find the optimal threshold,the proposed AdaBoost algorithm can not only adopt the proposed AdaBoost algorithm.The adaptive decision to implement physical layer access authentication can also improve the authentication rateof about 0.03.In order to improve the detection rate,a physical layer channel authentication scheme based on machine learning is proposed.Establish a two-dimensional feature-based authentication model,collect the actual USRP platform communication channel data,and verify the feasibility of the model by combining multiple machine learning classification algorithms.Compared with the manual ergodic threshold scheme,two-dimensional features,multiple machines The learning classification algorithm can significantly improve the detection rate,and the authentication ratecan be increased to 0.99,and the detection time is greatly reduced.The access authentication scheme becomes adaptive and intelligent,fully meeting the needs of real-time control access.In order to identify malicious nodes,a malicious node attack identification scheme based on edge computing is proposed.First,the identity of the unknown access node is determined to determine whether it is a clone node or a Sybil node.Furthermore,it is combined with the machine learning classification algorithm to train and learn,and the open source database data is used for verification.The results show that the detection rates are high in the four scenarios,making the identification scheme for malicious nodes more complete and adaptive.In summary,this paper takes edge computing as the application background,and mainly studies the access authentication technology under wireless communication,model establishment,algorithm derivation,algorithm example,simulation analysis,platform verification,and deepening,and has obtained certain research results.