Equilibrium Strategy of Attack and Defense in Computer Networks Based on Markov Signal Game Theory

In actual network attack and defense, the information between the two sides is often unbalanced and asymmetric, and the defender cannot fully understand the attacker’s true intentions and attack methods. At the same time, the complexity, rapid changes, and limited resources of network attack and defense make it crucial to optimize network defense strategies. The moving target defense strategy is introduced, and the moving target Markov signaling game defense strategy is built to optimize the defense decision. PageRank algorithm is applied to compute the stage weight value in the multi-stage process. The refined Bayesian equilibrium of the model is calculated. From the results, when the security resources were 8, the average defender profit was −0.95, which was 3.11 and 19.15 higher than that of the single-stage Stackelberg and the equitable distribution model, respectively. The average attacker profit was below the other models when the number of security resources was less than 10. Among them, when the number of security resources was 4, the average profit of attackers in the moving target Markov signaling game defense model was only 1.78, which was significantly lower than that of the single-stage Stackelberg model and the equitable distribution model. In addition, the detection/defense success rate proposed by the research is the highest. This proves that the model proposed in the study can effectively enhance the defense ability against network attacks, greatly improve network security defense technology, and provide new reference technologies for network security maintenance.