Research On Network Risk Defense Method Based On Multi-level Game Model

In recent years,with the frequent occurrence of network security incidents,people have paid more and more attention to network security.Researchers have given a variety of assessment and defense methods in the field of network risk assessment and defense,and have achieved certain results.However,from the practical application results of the method,there are still shortcomings in which network risk assessment and defense cannot be effectively combined.In view of the shortcomings of the existing research,this paper combines the traditional analytic hierarchy process and game theory.By constructing a multi-level game model to evaluate the degree of network risk,the value of each vulnerability threat in the network is obtained,and the benefits of both sides are attacked.The optimal defense strategy is obtained by defining and quantifying the benefits of both offense and defense.The main work of this paper is as follows:Firstly,based on the relevant indicators affecting network security,considering the complex hierarchical structure of the network,a multi-level game model is established.The model combines two methods,AHP and game theory,which can be able to propose proactive defense strategies and reduce network risks after network risk assessment.Secondly,in order to evaluate the network risk,this paper proposes an optimization method for the judgment matrix structure in the analytic hierarchy process by comparing the weight of the network risk degree between the network vulnerabilities and the different security indicators.As the construction basis of the judgment matrix in the network risk assessment,it partially compensates for the deficiency of the traditional analytic hierarchy process relying too much on expert experience.Thirdly,in order to select the active defense strategy,the profit of both sides of the network is defined and quantified by analyzing the defender's revenue.The gambit software package is used to calculate the optimal defense strategy under the current network.A judgment matrix update algorithm is proposed,which can destroy the attack path according to the defense strategy,reconstruct the judgment matrix and obtain new network risk weights.The proposed algorithm provides a basis for verifying the effectiveness of the defense strategy and can provide guidance for the selection of defense strategies.Finally,through experiments,the multi-level game model is used to evaluate the network risk and select the defense strategy,and verify the effectiveness of the defense strategy.