Research On Network Security Defense Decision-making Method Based On Time Game Model

Nowadays,as information technology developing fast,network applications follow that step and enable other industries to bloom.Rapid development of network technology and services accompanies with increasingly severe threat and unoptimistic security situation.It’s been a priority for network administrators to improve the comprehensive network security prevention and control system and comprehensively enhance the network security protection capability.The essence of network security is attack-defense confrontation,due to network attack-defense confrontation basic characteristics highly consistent with game theory.Appling game theory to turn real-world network security problems into game theory mathematical models to realize scientific and accurate modeling analysis and predictive deduction of network attack and defense process can provide theoretical tools for studying network defense decision-making problems.The research of network security defense decision-making method based on game model is one of the frontiers and hot spots in the field of network security.Network attack-defense confrontation analysis with game theory mainly focus 2 aspects.One is attack-defense behavior,or action strategy.The other one is opportunity,or time strategy.Current network defense decision-making method with game model mostly focus on behavior modeling.Configuring pertinent defense,adjusting defense order,and changing defense intensity are applied to improve network security defense efficacy.Those relative research are remarkable.But current models neglect time opportunity influence on network security.Lack of active selection on defense opportunity of best interest when facing abrupt and concealing network attack severely restricts defense efficacy.Besides,traditional network defense decision methods mostly adopt random network model to abstractly describe real network and neglect complex structure characteristics of real network.Practicability and universality need to be improved.This thesis starts with attack-defense confrontation to study behavior opportunity and intensity facing typical network topological structure of different features.Firstly,the classical network structure model and network propagation dynamics model are analyzed,and the research status and existing problems of network attack and defense game model and defense decision-making method are sorted out.Secondly,the stochastic network attack-defense time game model and the complex network attack-defense time game model are constructed to study the defense timing decision problem in different network structures and different strategic dimensions.Finally,the differential game model of network attack and defense time is constructed to study the unified decisionmaking problem of defense timing and defense action integration.The effectiveness and advancement of the proposed model and method are verified by simulation experiments,data statistics and comparative analysis.The main work includes the following aspects:1.Aiming current network attack-defense game model on attack-defense timing factor lack of consideration,the study proposes stochastic network defense opportunity decision-making with time game.This thesis Uses the propagation dynamics model to describe the evolution of the security state of network system resources changing in real time,construct a stochastic network attack and defense time game model based on Flip It game theory,analyze the network attack and defense confrontation process and quantify the attack and defense benefits in terms of the length of time both sides control the target resources from the perspective of attack and defense timing;According to the game equilibrium design network defense time decision algorithm,determine the optimal defense time strategy.Through simulation experiments and comparative analysis,it is verified that real-time and dynamic adjustment of defense strategy can improve defense revenue,which further illustrates the importance of attack and defense action timing.The research results provide theoretical support for the research on the optimal defense timing decision method for complex networks.2.The network in real life is a complex network with small-world effect and scale-free characteristics.Considering the influence of network structure characteristics on the network security state,the study proposes a complex network defense decision method based on time game.Aiming at the characteristics of different environments and different decision-making behaviors of each node in complex networks,the evolution differential equation of complex network node state is constructed to describe the evolution process of network security state.Construct a complex network attack and defense time game model,quantify the attack and defense income function,solve the game equilibrium,and then design the optimal defense time strategy selection algorithm;finally,a small-world network and a scale-free network experimental environment are built for verification,and the defense effectiveness analysis and defense effectiveness comparison analysis are carried out.The results show that the complex network structure can have a greater impact on network security.At the same time,it is concluded that dynamically adjusting the defense timing can effectively improve the network defense effectiveness,which verifies the validity and scientificity of the model and method.3.For a network security defense opportunity and behavior lack of unified decision method problem,this thesis proposes stochastic network unified decision-making method with timedifferential game.Network security status depends on confrontation intensity and opportunity.The key to enhance network defense efficacy is to study defense behavior and opportunity unified method best combination point.This thesis systematically analyzes stochastic attack-defense behavior and opportunity,referring to propagation dynamics model realizing behavior and time strategy combination on network security status.Unify time game with differential game theory to construct stochastic network attack-defense time game differential model to quantify attackdefense revenue from time and space angles.On solving game saddle point equilibrium to design network defense decision method and determine optimized unified defense strategy.Simulation experiment verify our method to improve network defense efficacy on intensity and opportunity to unify ensemble defense strategy.