基于SVM的VANET虚假消息检测机制

SVM-Based False Message Detection Mechanism in VANET

VANET依赖于节点间频繁的信息交互,消息的真实性和时效性至关重要。传统的身份认证和加密机制只能确保用户身份合法,但无法识别内部不端节点。内部不端节点可能发送虚假的交通告警信息来干扰其他用户的正常驾驶,严重时造成人员伤亡。针对虚假消息攻击,本文通过提取交通告警信息中的多维特征,采用支持向量机(SVM)来实现消息的识别与分类,仿真结果表明:SVM在VANET交通信息分类方面具有优良性能,在小样本情况也具有较好的泛化性能,实现了较高的虚假消息的检测率。

VANET relies on frequent information interaction between nodes,and the authenticity and timeliness of the message is very important.Traditional authentication and encryption mechanisms can only ensure that the user is legal but can not identify the internal misbehavior node.The internal misbehavior node may send false traffic alarm information to interfere with the normal driving of other users,and even cause serious casualties.In this paper,SVM is used to identify and classify the message by extracting the multidimensional features in the received alarm information.The simulation results show that SVM has excellent performance in VANET traffic classification and has good generalized performance even in small samples,achieving a high false alarm detection rate.