Research On Privacy Protection Of Internet Of Things Based On Game Theory

With the rapid development of the technology and application of the Internet of Things,the Internet of Things is playing an increasingly important role in the home and business fields,and has made great contributions to improving the quality of life and promoting the growth of the world economy.However,due to the huge amount of data in the Internet of Things and the wireless communication mode adopted by most Internet of Things applications,there are potential privacy threats in the Internet of Things.Firstly,compared with traditional computers,devices in the sensing layer of the Internet of Things are more vulnerable to attacks.Especially in some cases,users are not aware that the data was collected by smart devices in small homes,which contain more privacysensitive data.Moreover,devices in the sensing layer have limited resources,making it difficult to implement energy-consuming security systems.Secondly,as the basis of the Internet of Things,a sensor network is widely distributed and lacks supervision,which makes it difficult to guarantee the security of each node.Therefore,nodes are easy to be captured by attackers.And the application layer processes the data collected and provides it to users,and the process also requires privacy protection measures.All these situations put forward higher requirements for privacy protection on the Internet of Things.Game theory provides many tools and methods for solving competition and interaction problems among participants,which are well suited to the analysis,modeling,and decision-making process of privacy.In order to solve the privacy protection challenges in the process of information sharing on the Internet of Things,this paper aims at maximizing the privacy protection and minimizing the availability of the Internet of Things,and builds a privacy protection model of the Internet of Things based on the game theory.The main research contents are as follows.In view of the problem that nodes maliciously request resources of the Internet of Things,and considering the characteristics of bounded rationality of both sides,a privacy protection model of the Internet of Things is proposed based on an evolutionary game.Information entropy is used to define the privacy risk factor,and it is introduced into the privacy protection model of the Internet of Things to describe the risk level of node requests.Then,the evolutionary stability strategy equilibrium is calculated by replicating the dynamic equation,and the Jacobian matrix is used to analyze the evolutionary stability strategies of the participants in the privacy protection model under different conditions.On the other hand,trust is introduced to reward nodes for taking normal requests.The simulation results show that the trust does have an incentive effect on the normal request strategy of nodes,and the smaller the privacy risk factor,the more likely the nodes are to take the normal request,which provides a reference for whether the Internet of Things resource system authorizes the node request.Aiming at the Internet of Things with malicious nodes,a privacy protection model based on signaling game is proposed.On the basis of considering detection rate and false alarm rate,a more complete payment matrix is constructed by introducing the parameters such as trust degree and attack lethality,and the privacy is measured by using information entropy.The node belief was updated according to the Bayesian principle,then the equilibrium strategy in multi-stage dynamic game was analyzed,the optimal node detection strategy and attack strategy were sought,and the influence of different strategies on privacy protection of the Internet of Things was studied.The simulation experimental results show that the detection nodes which improve detection rate and reduce the rate of false positives can effectively improve the degree of privacy,the optimal attack probability of malicious nodes decided jointly by the fatal attack and the rate of false positives,and the influence of attack lethality on the optimal attack probability of malicious nodes is more significant,which provides a theoretical basis for the design of privacy protection mechanism of the Internet of Things.