Scholarlink

Software-Defined Networking (SDN) centralizes and programs network management by decoupling control and data planes, offering enhanced visibility compared to traditional networks. Its flexibility, scalability, and security make it increasingly adopted in healthcare, where patient data integrity is critical. However, SDN-based healthcare systems remain vulnerable to cyber threats like DDoS, scanning, and injection attacks. This paper proposes MCAD, a Machine Learning-based Cyberattack Detector for SDN-enabled healthcare. MCAD customizes an L3 switch application under the Ryu controller to monitor and classify data flows, enabling real-time threat detection. A comprehensive evaluation benchmarked various ML algorithms against cyberattacks, providing a comparative analysis of their strengths and limitations. These findings contribute to proactive cybersecurity in SDN-enabled healthcare networks.

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

Intelligent Monitoring and Threat Classification: A Machine Learning-Driven Framework for Enhancing Cybersecurity in SDN-Based Healthcare Networks

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shaista Tarannum,
  • A. P. Jyothi,
  • Sri. H. Balaji,
  • Veerbhan Patel,
  • K. N. Jayaprasad,
  • Shashank S. Kumbar,
  • S. R. Bhagyashree,
  • Shreyanka Subbarayappa,
  • N. Bhavatarini

摘要

Software-Defined Networking (SDN) centralizes and programs network management by decoupling control and data planes, offering enhanced visibility compared to traditional networks. Its flexibility, scalability, and security make it increasingly adopted in healthcare, where patient data integrity is critical. However, SDN-based healthcare systems remain vulnerable to cyber threats like DDoS, scanning, and injection attacks. This paper proposes MCAD, a Machine Learning-based Cyberattack Detector for SDN-enabled healthcare. MCAD customizes an L3 switch application under the Ryu controller to monitor and classify data flows, enabling real-time threat detection. A comprehensive evaluation benchmarked various ML algorithms against cyberattacks, providing a comparative analysis of their strengths and limitations. These findings contribute to proactive cybersecurity in SDN-enabled healthcare networks.