| Wireless Sensor Networks(WSNs)are the integration of Micro-Electro-MechanicalSystems and The Internet of Things systems.They have the characteristics of limited resource level,large node scale,data-centricity,self-organization and wide application.City,smart medical,industrial control,smart home and environmental monitoring and many other fields.With the integrated development and wide application of Fifth Generation Mobile Communication Technology(5G)and wireless sensor networks,WSNs are increasingly subject to security attacks and face numerous security risks.Network security management issues have become obstacles to their development.In order to improve the security performance of WSNs and reduce the threat brought by network attacks,this paper conducts security management research from three aspects: authentication of device access at the perception layer of WSNs,intrusion detection algorithm at the network layer,and secure transmission of user data at the application layer.The main work is as follows:1.Aiming at the resource-constrained WSNs perception layer devices,group access authentication is adopted to reduce the network signaling bearer,and a resource-constrained wireless sensor device access authentication selection mechanism oriented to the perception layer is proposed.Firstly,the wireless sensor network equipment adopts the 5G network access authentication protocol,which leads to low device authentication efficiency,so group authentication methods such as group leader authentication and proxy authentication are used;Secondly,analyze the resource overhead of equipment using group leader and proxy authentication;Finally,based on the RFC7228 standard for three types of resource constrained devices issued by the IETF working group,and considering WSNs’ considerations on network deployment factors such as device deployment cost,network security and network signaling bearer,a resource constrained device access authentication selection mechanism is proposed.The experimental results and data analysis show that the access authentication selection mechanism meets the resource level and deployment requirements of WSNs devices,and greatly reduces the number of network signaling.2.Aiming at the different requirements of WSNs for deploying malicious encrypted traffic intrusion detection algorithms in the network layer in different application scenarios,a network layer-oriented WSNs intrusion detection algorithm matching management is proposed.Firstly,seven typical malicious encrypted traffic intrusion detection algorithms are introduced,and the structure and characteristics of the algorithms are analyzed;Secondly,based on the WSNs application scenarios for the different requirements of intrusion detection algorithm classification accuracy,computing resource occupancy and storage resource occupancy,a management scheme for intrusion detection algorithm allocation is proposed;Finally,based on the transport layer security protocol,the public data set CICIDS-2017 is selected to generate a two-category encrypted traffic data set,and seven intrusion detection algorithms are trained.The experimental data show that the long short-term memory network algorithm has the highest classification accuracy among the seven algorithms;the random forest algorithm occupies the lowest storage resource;the naive Bayesian algorithm,the quadratic discriminant analysis algorithm,the random forest algorithm and the adaptive boosting algorithm perform better in computing resource occupation.Therefore,the proposed intrusion detection algorithm with management scheme is in line with the analysis of the experimental results.3.Aiming at 5G communication-oriented WSNs ignoring the application layer user data security transmission requirements in virtual resource scheduling,a virtual resource scheduling model based on application layer data security transmission level and user historical priority is proposed.Firstly,the characteristics of two typical resource scheduling models based on user request sequence and user historical priority are analyzed.The users of application layer services are network slice users,WSNs are mobile virtual operators,and the underlying physical resources that provide 5G communication guarantees are the basic service providers;Thirdly,the user’s resource allocation is determined based on the user’s data security transmission requirements and the user’s historical resource allocation.In order,a virtual resource scheduling model based on application layer data security transmission level and user historical priority is proposed;Finally,code is written on the simulation platform to simulate the experiment.The experimental results show that,compared with the other two models,the application-layer-based data security transmission level and user historical priority virtual resource scheduling model can meet the needs of user data security transmission,effectively increase the cost of malicious user side channel attacks and increase the mobile virtual operator’s cost income. |
