基于信任距离的车联网恶意节点检测方法

发布时间：2018-04-06 23:35

本文选题：车联网　切入点：信任模型　出处：《计算机科学》2015年08期

【摘要】：针对车联网网络拓扑结构变化快且相比传统移动自组网络更易受到恶意车辆发起的内部攻击等问题,在当前贝叶斯假设的信任理论研究的基础上,结合车联网高速移动中快速检测恶意节点的要求,加大否定事件的影响力度,提出了用于评估车辆节点行为的信任模型;在综合推荐信任值时,引入了"推荐信任距离"作为推荐信任的信任度量,预先排除恶意推荐意见,并有效防止车辆的串通攻击。与现有的基于信任的检测方法相比,该方法加快了检测速度,并简化了推荐传递。仿真实验表明,该方法有较快的检测速度,从网络丢包率和恶意节点检测率可以看出此信任模型对检测恶意节点具有较好的性能。
[Abstract]:Aiming at the problem that the network topology changes rapidly and is more vulnerable to the internal attack by malicious vehicles than the traditional mobile ad hoc networks, based on the current trust theory research of Bayesian hypothesis.Combined with the requirement of fast detection of malicious nodes in high-speed mobile vehicle network, the influence of negative events is increased, and a trust model for evaluating the behavior of vehicle nodes is proposed, and when the trust value is synthetically recommended, a trust model is proposed to evaluate the behavior of vehicle nodes.The recommended trust distance is introduced as the trust degree of the recommendation trust, and the malicious recommendation is excluded in advance, and the collusion attack of the vehicle is effectively prevented.Compared with the existing trust-based detection methods, this method speeds up the detection speed and simplifies the transfer of recommendations.The simulation results show that the proposed method has a high detection speed. From the network packet loss rate and the malicious node detection rate, it can be seen that the trust model has good performance in detecting malicious nodes.
【作者单位】：江苏大学计算机科学与通信工程学院;
【基金】：国家自然科学基金(61272074) 江苏省自然科学基金(BK2011464) 镇江市工业支撑计划项目(GY2013030)资助
【分类号】：TP393.08;TN929.5

【参考文献】