Adaptive and Lightweight DDoS Defense in SDN-Enabled Edge–IoT Networks Using Unsupervised and Reinforcement Learning

View at publisher Summarise with VIRAT
Authors: Ananth Kumar R, Ramachandra Reddy, Bharath Kumar S
Cited by: 0
References: 20

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

The high rate of growth in the Internet of Things (IoT) devices within the edge computing settings has grown the vulnerability of the networks to Distributed Denial of Service (DDoS) attacks exponentially and necessitates efficient, adaptive, and defensive solutions. In this work, a novel SDN-Edge-IoT-DDoS detection and mitigation system is suggested which makes use of an unsupervised-powered anomaly detection model along with a reinforcement learning-based countermeasures in mitigation strategies. In contrast to the traditional methods based on a fixed threshold or signature-like detection, the suggested solution constantly hears traffic patterns at the Software-Defined Networking (SDN) controller and changes mitigation policies dynamically with the current network behavior and predicts potential attack vectors before they cause a substantial disruption. The framework uses a lightweight feature extraction that is adapted to resourceconstrained IoT devices and thus it has minimal overhead. The results of simulation prove that the methodology positively influences detection accuracy, false positives and resource amount usage, creating a self-learning, context-aware, and predictive security paradigm of SDN-enabled edge-IoT networks. This represents a novel integration of predictive Machine learning (ML)-based mitigation being integrated into an SDN-edge-IoT ecosystem.

Fields of Study & Subjects

Network Security and Intrusion Detection Software-Defined Networks and 5G Network Packet Processing and Optimization

Publication details

Journal / source
International Conference on Intelligent Data Communication Technologies and Internet of Things (IDCIoT)
Pages
476–482
Publication date
4 February 2026
Language
EN