基于最优反应均衡的传感网恶意程序传播抑制方法

Quantal Response Equilibrium-Based Method for Preventing WSNs Malware Infection

为抑制传感网恶意程序传播,在考虑传感网恶意程序传播参与者"有限理性"的基础上,提出一种基于最优反应均衡的方法。根据传感网恶意程序传播过程中的博弈分析,建立传感网恶意程序传播阶段博弈模型以反应传感网恶意程序和传感网入侵检测系统之间的博弈交互过程。由参与者之间博弈交互持续进行的事实,建立传感网恶意程序传播重复博弈模型。使用最优反应均衡预测传感网恶意程序的行为以解决重复博弈纳什均衡解求解困难的问题,给出抑制传感网恶意程序传播的算法。实验分析了参与者基于最优反应均衡的策略,对所提出方法的有效性进行了验证。

We consider bounded rationality of players during the process of WSNs（ Wireless Sensor Networks） malware infection,and propose a method based on QRE（ Quantal Response Equilibrium） to prevent the infection behavior of malware. According to game analyses,we construct a stage game model to reflect interactions between two players-WSNs malware and WSNs IDS（ Intrusion Detection System）. Furthermore,we construct a repeated game model describing continual interactions between the two players. We then solve the problem of computing Nash Equilibrium in the repeated game by employing QRE to predict behaviors of WSNs malware,and attain an algorithm preventing WSNs malware infection. Experiments analyze QRE-based strategies for the two players,and confirm the efficiency of our method.