Research On The Key Trusted Technology For The Perception Layer Of The Internet Of Things

Compared with the traditional Internet,mobile intelligence computing has become a support for the Internet of Things(IOT),which making its security issues more complicated than the Internet.Especially on perception layer of the IOT,large numbers of heterogeneous sensing networks,such as security networking,safe operation,data security transmission and malicious behavior detection and so on,make the security problem even more prominent,and hence the IOT perception layer is the key layer for the trusted operation of the IOT.To ensure the reliable operation of the sensing node,what kind of security mechanism is adopted has a great significance in improving the credibility of the IOT infrastructure.(1)In view of the lack of trusted running mechanism of sensing nodes,a trusted running mechanism is proposed,which combines the identity authentication of the nodes,the initial state measurement,and the behavior attribute measurement.In this mechanism,when the sensing node carries out data acquisition and transmission in the sensing network,the superior node needs to implement identity authentication and initial state measurement to ensure the identity and initial state of the node to be trusted.At the same time,a behavioral confidence measurement model of perceived nodes is proposed.First,the model uses a behavioral Trust Base Library to make a preliminary decision,and then uses a convolution neural network for data mining to detect unconventional behavior.Through experimental simulation,it is found that the model proposed in this paper has good applicability for the sensing nodes with limited computing power.It can realize the dynamic confidence measurement of the sensing nodes,the trusted measurement value is fast and accurate,and has good active control characteristics at the same time.(2)In view of the lack of continuous integrity measurement mechanism,the sensing layer node has a fine-grained property measurement mechanism.In view of this problem,a trusted connection architecture TCA-BAM is proposed to realize the bidirectional identity authentication and the integrity identification of the dual platform and the IOT information of the access point and the sensing network node of the IOT.The center identifies trusted behaviors and attributes of the sensing network.At the same time,a sening layer network trusted access scheme based on behavior attributes is proposed.The scheme completes the bidirectional identity authentication,platform credibility authentication and behavior attribute authentication of the cluster head nodes to the perceived nodes by the key allocation scheme based on symmetric algorithm.Experiments show that the mechanism is measurable,monitorable and extensible,and can effectively confirm the credibility of data sources and improve the availability of the system.(3)The author also conceives a data trusted transmission scheme based on compressed sensing to fullfill security requirements for data compression transmission.The scheme improves the accuracy of data recovery by using the AR model and the spatial sparsity of data.It ensures both the low energy consumption of data transmission and the safe transmission of data.Compared with existing methods,the scheme has better data reconstruction results.(4)In view of the problem that most terminals in the Internet of things do not have a high security configuration due to the constraints of the environment,a user identity authentication scheme based on dynamic gestures is proposed.In this scheme,a dynamic gesture model based on multi instance is proposed.Based on the model,a dynamic gesture based identity authentication mechanism of Internet of things is proposed,and the identity authentication between the users of the Internet of things is realized by building the user dynamic gesture Library of the Internet of things.Experimental analysis shows that the scheme has high gesture recognition rate and high security user identity authentication.