In the network attack-defense game systems

there are many stochastic factors

such as changes of attack-defense strategy sets and system operating environment. The traditional deterministic game model can not describe the game process of network attack and defense accurately. This paper constructed an attack-defense stochastic evolutionary game model by adapting the nonlinear Itó stochastic differential equations. The model can be applied to analyze the stochastic evolutionary process of network attack and defense. In addition

the stability of the strategy selection of attack and defense was analyzed according to the discriminant theorem of stochastic differential equations. Besides

an algorithm to select the security defense strategies based on stochastic attack-defense evolutionary game model was designed. Finally

the simulations demonstrate that the different intensity influences of stochastic interference on the speed of decision-making evolution of attack and defense. The attackers and defenders are more inclined to choose strong strategies when the game system has great intensity of interference. The model and the method proposed in this paper can provide guidance for attack behavior prediction and defense strategy selection.