论文部分内容阅读
针对传统攻击流量的集中式检测模型中可扩展性差,检测效率低以及误报率高等问题,设计了针对DDoS攻击流量的随机森林分布式检测模型,该模型包括数据采集模块、数据预处理模块、分布式分类检测模块和报警响应模块.将该模型与基于Adaboost算法的分布式检测方法进行比较,并通过实验研究验证了模型的有效性.结果表明:基于随机森林的组合分类器分布式检测模型具有更高的检测率、正确率、精确率以及更低的误报率,并且该模型部署灵活,适用于工程实践.
Aiming at the problems of poor scalability, low detection efficiency and high false alarm rate, a distributed forest detection model based on DDoS attack traffic is proposed for the centralized detection model of traditional attack traffic. The model includes data acquisition module, data preprocessing module, Distributed classification detection module and alarm response module.This model is compared with the distributed detection method based on Adaboost algorithm and the validity of the model is verified through experimental research.The results show that the distributed classifier detection model based on random forest With a higher detection rate, accuracy, accuracy and lower false alarm rate, and the flexible deployment of the model, suitable for engineering practice.