DoS Attack Detection Based on Deep Factorization Machine in SDN

作     者：Jing Wang Xiangyu Lei Qisheng Jiang Osama Alfarraj Amr Tolba Gwang-jun Kim 

作者机构：School of Computer&Communication EngineeringChangsha University of Science&TechnologyChangsha410114China Computer Science DepartmentCommunity CollegeKing Saud UniversityRiyadh11437Saudi Arabia Department of Computer EngineeringChonnam National UniversityGwangju61186Korea 

出 版 物：《Computer Systems Science & Engineering》 (计算机系统科学与工程（英文）)

年 卷 期：2023年第45卷第5期

页      面：1727-1742页

核心收录：

学科分类：0839[工学-网络空间安全] 08[工学] 081201[工学-计算机系统结构] 0812[工学-计算机科学与技术（可授工学、理学学位）] 

基　　金：This work was funded by the Researchers Supporting Project No.(RSP-2021/102)King Saud University,Riyadh,Saudi Arabia This work was supported by the Research Project on Teaching Reform of General Colleges and Universities in Hunan Province(Grant No.HNJG-2020-0261),China 

主　　题：Software-defined network denial-of-service attacks deep factorization machine GRMMP 

摘      要：Software-Defined Network(SDN)decouples the control plane of network devices from the data *** alleviating the problems presented in traditional network architectures,it also brings potential security risks,particularly network Denial-of-Service(DoS)*** many research efforts have been devoted to identifying new features for DoS attack detection,detection methods are less accurate in detecting DoS attacks against client hosts due to the high stealth of such *** solve this problem,a new method of DoS attack detection based on Deep Factorization Machine(DeepFM)is proposed in ***,we select the Growth Rate of Max Matched Packets(GRMMP)in SDN as detection ***,the DeepFM algorithm is used to extract features from flow rules and classify them into dense and discrete features to detect DoS *** training,the model can be used to infer whether SDN is under DoS attacks,and a DeepFM-based detection method for DoS attacks against client host is *** results show that our method can effectively detect DoS attacks in *** with the K-Nearest Neighbor(K-NN),Artificial Neural Network(ANN)models,Support Vector Machine(SVM)and Random Forest models,our proposed method outperforms in accuracy,precision and F1 values.