Detection and mitigation of DDoS attacks in internet of things using a fog computing hybrid approach

Abstract

The introduction of a new technology has aided the exponential growth of the internet of things (IoT), allowing for the connecting of more devices in the IoT network to be made possible by the availability of quicker connections and reduced latency. As IoT networks have become more prevalent and widely used, security has become one of the fundamental requirements, and a distributed denial of service (DDoS) attack poses a significant security threat due to the limited resources (CPU, memory, open source, persistent connection) that can be used to either intentionally or unintentionally increase DDOS attacks. Fog computing is proposed in this study as a framework for real-time detection and mitigation of DDoS assaults. Fog computing is accurate and quick in detecting attacks due to its proximity to IoT devices. DDOS assaults are detected using an approach that combines randomness measurement of traffic with k-nearest neighbors (KNN) machine learning algorithm. Suggested system obtained 100% detection accuracy for transmission control protocol TCP attacks, 98.79% detection accuracy for UDP attacks, and 100% detection accuracy for internet control message protocol ICMP attacks.