Abstract: The evaluation mechanism based on the trust degree of network nodes is an effective method to prevent worm propagation. Aiming at the problem that the existing node trust degree calculation model is insensitive to worm attack behavior and not comprehensive to the identification of malicious nodes, this paper proposes a trust update model that considers the risk of worm propagation. Through the definition of the STIDR state in the worm propagation model, the relationship between node trust degree and propagation probability is quantitatively characterized. And a trust update model is established under three conditions of direct risk, indirect risk and propagative risk. The simulation results show that compared with the basic trust model, the trust update model considering the risk of worm propagation in this paper makes the trust degree of infected state nodes lower, the number of infected state nodes less, and the number of isolated state nodes more, which indicates the update of the trust degree makes the identification of the infected state node more effective. To a certain extent, the purpose of resisting worm attacks and inhibiting worms from spreading in the network is achieved.

Key words: worm propagation;trust model, trust degree, propagation probability

摘要： 基于网络节点信任度的评估机制是一种有效防御蠕虫传播的方法。针对现有节点信任度计算模型对蠕虫攻击行为不敏感,对恶意节点识别不全面的问题,本文提出了一种考虑蠕虫传播风险的信任度更新模型。通过蠕虫传播模型中STIDR状态定义,对节点信任度和传播概率的关系进行了定量刻画;建立了直接风险、间接风险和传播性风险三种条件下的信任度更新模型。仿真结果表明,与基本信任模型相比,本文考虑蠕虫传播风险的信任度更新模型使网络中感染状态节点的信任度更低,感染状态节点数目更少,隔离状态节点数目更多,表明信任度的更新使节点对感染状态节点的识别更加有效,在一定程度上达到了抵御蠕虫攻击,抑制蠕虫在网络中传播的目的。

关键词: 蠕虫传播, 信任模型, 信任度, 传播概率