Research On Location And Identity Privacy Protection In Fog Computing

With the number of network terminals gradually increasing,the centralized computing model which is represented by cloud computing can not deal with the massive data witch is generated at the edge of the network.The fog computing solves the problems in system response,delay,and bandwidth.Traditional data security and privacy protection technologies require devices to have strong storage and computing capabilities.However,the capabilities of the fog computing node are not strong enough.Therefore,the existing protection technologies cannot be fully applied to the fog computing model.The problems of data security and privacy protection in fog computing are more prominent.In summary,the research on privacy protection in the fog computing model has important practical significance.Firstly,this paper analyzes the current main privacy protection technologies,which are mainly divided into three types: policy method,distortion method and encryption method.This paper proposes a location and identity privacy protection scheme for fog computing in hierarchical management of pseudonyms based on existing research.The scheme divides the system model into a cloud layer,a fog layer,and a terminal layer and moves the pseudonymous production and management agencies from cloud layer to fog layer.The operation process of the scheme can be divided into 6 protocols.And the profit function which can be used to guide pseudonymous production of pseudonym production is obtained according to the three factors of pseudonym profit,pseudonym fog loss and pseudonym storage overhead.This paper introduces identity certificates and pseudonym certificates,which effectively solves the problems of pseudonym replacement and cross-domain use of pseudonyms.Secondly,this paper designs a set of information entropy models for fog computing which are based on the theory of information entropy.And the information entropy models has multi-privacy sources with subjective feelings and privacy attacks.The privacy condition entropy in the model represents the protection effect of the privacy protection scheme.The private mutual information in the model represents the attacker's attack capability.The pseudonym replacement policy based on information entropy and context awareness considers three factors: the node's privacy information entropy,pseudonym replacement cost,and the environment where the edge node is located.The network edge node decides whether to replace the pseudonym group in theinvisible area based on these three factors.Finally,this paper takes smart transportation as an example to compare hierarchical management of pseudonym scheme with the traditional centralized pseudonym management scheme.The results show that the delay is reduced by 60% and the communication overhead of pseudonym replacement request reduced by 50%.The analysis in the optimal pseudonym production scheme shows that the pseudonym profit and the pseudonym fog loss play a positive role,and the pseudonym storage cost plays a negative role.In the analysis of pseudonym replacement strategies based on information entropy and context awareness,the correctness of the information entropy model is verified.Then this paper analyzes the influence of marginal variable conditions,the cost of pseudonym replacement and the environment where the edge crosses on the pseudonym replacement intention.The final result verified the efficiency,feasibility and correctness of the scheme.