基于攻防信号博弈模型的防御策略选取方法

Defense policies selection method based on attack-defense signaling game model

当前基于博弈理论的防御策略选取方法大多采用完全信息或静态博弈模型,为更加符合网络攻防实际,从动态对抗和有限信息的视角对攻防行为进行研究。构建攻防信号博弈模型,对策略量化计算方法进行改进,并提出精炼贝叶斯均衡求解算法。在博弈均衡分析的基础上,设计了最优防御策略选取算法。通过实验验证了模型和算法的有效性,并在分析实验数据的基础上总结了攻防信号博弈的一般性规律,能够指导不同类型防御者的决策。

Currently defense policies selection based on game theory mostly applied either the complete information game model or the static game model. In order to be more in line with the reality of network attack and defense, attack-defense behavior was studied by dynamic rivalry and incomplete information. The attack-defense signaling game mode was built, the method to quantify policies was improved and an algorithm to obtain the perfect Bayesian equilibrium was proposed. On the basis of analyzing equilibrium, the algorithm for selecting the optimal defense policy was proposed. The simulation experiment demonstrates that the model and algorithms are feasible and effective. By the experimental data, general rules on signaling attack-defense game are summarized, which can guide defenders of different types to make decisions.