Scholarlink

<p>Distributed denial-of-service (DDoS) attacks have become an increasingly serious and growing threat to modern network infrastructures, with a significant increase in frequency, sophistication, and scale. Despite providing dynamic programmability and centralized control, software-defined networks (SDNs) are not immune to these threats. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), this systematic literature review examines the state of the art of machine learning, deep learning, and hybrid techniques utilized to detect and prevent DDoS attacks in SDNs between 2019 and 2024. This review aimed to identify the most widely deployed ML and DL techniques for DDoS detection in SDN, pinpoint active research fields, explore the practical challenges of implementing these systems in various SDN environments, and identify future directions for SDN security against DDoS. We analyzed 62 articles from major databases (IEEE Xplore, ScienceDirect, Scopus, SpringerLink, Web of Science) under the PRISMA protocol. The results of our analysis revealed a gradual increase in publications during the studied period, proving growing interest in the importance of DDoS attack detection and prevention in SDN environments. However, the authors identified various limitations in the existing studies related to the learning process, the use of inappropriate datasets, limited evaluation metrics, and insufficient emphasis on DDoS attacks. In addition, various implementation issues related to scalability, real-time processing, and integration of smart DDoS attack detection systems in SDN architectures. The authors believe that this review will help researchers propose more accurate DDoS attack detection and prevention systems to ensure SDN security.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Mitigating DDoS attacks in software-defined networks: a systematic literature review of machine learning and deep learning approaches

  • 期刊论文

  • 发表日期：

    2025年12月27日

查看全文
  • Kaoutar Tebbaa,
  • Oumaima Chakir,
  • Yassine Maleh,
  • Mustapha Belaissaoui

摘要

Distributed denial-of-service (DDoS) attacks have become an increasingly serious and growing threat to modern network infrastructures, with a significant increase in frequency, sophistication, and scale. Despite providing dynamic programmability and centralized control, software-defined networks (SDNs) are not immune to these threats. Using the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA), this systematic literature review examines the state of the art of machine learning, deep learning, and hybrid techniques utilized to detect and prevent DDoS attacks in SDNs between 2019 and 2024. This review aimed to identify the most widely deployed ML and DL techniques for DDoS detection in SDN, pinpoint active research fields, explore the practical challenges of implementing these systems in various SDN environments, and identify future directions for SDN security against DDoS. We analyzed 62 articles from major databases (IEEE Xplore, ScienceDirect, Scopus, SpringerLink, Web of Science) under the PRISMA protocol. The results of our analysis revealed a gradual increase in publications during the studied period, proving growing interest in the importance of DDoS attack detection and prevention in SDN environments. However, the authors identified various limitations in the existing studies related to the learning process, the use of inappropriate datasets, limited evaluation metrics, and insufficient emphasis on DDoS attacks. In addition, various implementation issues related to scalability, real-time processing, and integration of smart DDoS attack detection systems in SDN architectures. The authors believe that this review will help researchers propose more accurate DDoS attack detection and prevention systems to ensure SDN security.