In this paper, we develop a novel game theoretic model of the interactions between an EDoS attacker and the defender based on a signaling game that is a dynamic game of incomplete information. We then derive the best defense strategies for the network defender to respond to the EDoS attacks. That is, we compute the perfect Bayesian Nash Equilibrium (PBE) of the proposed game model such as the pooling PBE, separating PBE and mixed strategy PBE. In the pooling equilibrium, each type of the attacker takes the same action and the attacker's type is not revealed to the defender, whereas in the separating equilibrium, each type of the attacker uses different actions and hence the attacker's type is completely revealed to the defender. On the other hand, in the mixed strategy PBE, both the attacker and the defender randomize their strategies to optimize their payoffs. Numerical illustration is also presented to show the efficacy of the proposed model.

在本文中，我们基于不完全信息动态博弈的信号博弈，开发了一种新的 EDoS 攻击者和防御者之间交互的博弈论模型。然后，我们得出网络防御者应对 EDoS 攻击的最佳防御策略。也就是说，我们计算了所提出的博弈模型的完美贝叶斯纳什均衡 (PBE)，例如池化 PBE、分离 PBE 和混合策略 PBE。在池化均衡中，每种类型的攻击者采取相同的动作，攻击者的类型不会被暴露给防御者，而在分离均衡中，每种类型的攻击者使用不同的动作，因此攻击者的类型完全被暴露给防御者. 另一方面，在混合策略 PBE 中，攻击者和防御者都随机化他们的策略以优化他们的收益。还提供了数值说明以显示所提出模型的有效性。