Study On Network Security Attack And Defense Strategy Model Based On Envolutionary Game

The security of the network and the privacy of users have aroused great attention with the increasing complexity of the network information system.It is particularly important to find new technologies that can protect network security,analyze and predict network attack and defense forms.Only relying on passive defense can not meet the needs of network security because of the rapid change of the network environment.Therefore,game theory has been gradually applied to the research of network attack and defense because some scholars found that the process of network attack and defense has something in common with the process of the game.The evolutionary game breaks through the limitation of the complete rationality of traditional game theory as a theory combining game theory with a dynamic evolution process.It has attracted the attention of researchers because it can better describe the dynamic process between groups evolving over time.In view of this,based on the evolutionary game theory of incomplete rationality,this thesis analyzes and forecasts the network attack and defense behavior strategy,and then obtains the optimal defense strategy.It effectively reduces the loss of the defenders and curbs the attack behaviors of the attackers.Specifically,this thesis includes the following two study contents.(1)The evolutionary game model of attack and defense based on the incentive mechanism is proposed.This thesis analyzes the network environment,constructs the network attack and defense scene,and introduces the incentive mechanism to build the model based on the punishment mechanism.The evolutionary stability equilibrium of the model is analyzed by using the replicated dynamic equation and the local stability rule of the Jacobian matrix.Network attack and defense groups gradually tend to a stable state through continuous evolution in different problem situations.And the defenders finally obtain the optimal defense strategy in this situation by constantly modifying and improving its behavior.In addition,the defenders invest defense strategy more and more active because the third-party regulatory authorities to implement incentive mechanism management for the players.Moreover,it can be foundthat the defense strategies invested by the defenders will lead to longer attack time and higher attack cost in the process of analyzing the evolution rule of different attack groups,so as to curb the enthusiasm of the attackers and facilitate the construction of a more secure and stable network environment.Finally,it can be intuitively found that the incentive mechanism can effectively improve the defenders' payoffs by comparing defense investment return introduced in this thesis.Applying it to the actual network environment,the model can predict and detect the attackers in time,and provide a certain basis for the selection of the best defense strategies,which has a certain positive effect on the protection of the network environment.(2)In this thesis,a stochastic differential game in network attack and defense is proposed.The model has Markov property and can describe the stochastic interference factors in the actual network environment.In the process of building the model,the harm degree of the attack strategies is fully considered by introducing the hazard analysis.The evolutionary equilibrium solution and stability of the model are proved by the related theoretical knowledge of stochastic differential equation.In different problem situations,the evolutionary analysis of players' strategy selection in the game is carried out through the Explicit Euler numerical method.Under the influence of stochastic interference factors,the model proposed in this thesis can reach a stable state and obtain the optimal defense strategy in different problem situations.Furthermore,it can be concluded that the defense investment return of the model proposed in this thesis is higher by introducing the defense investment return to compare with other literature.The strategy selections of attackers and defenders groups are more suitable for the actual situation of the network and have a certain role in promoting the protection of network security.