A framework for hello flood attack detection and mitigation in resource constrained RPL-Based IoT Networks

Abstract

The Internet of Things (IoT) and the Industrial Internet of Things (IIoT) have revolutionized modern industries, but the security of these networks remains a critical concern. The Routing Protocol for Low-Power and Lossy Networks (RPL), widely used in IoT and IIoT for routing, is vulnerable to various attacks, including Hello Flood (HF) attacks. These attacks exploit the resource-constrained nature of IoT devices, leading to increased energy consumption and rapid memory depletion. This dissertation investigates the impact of Hello Flood attacks on resourceconstrained IoT nodes, particularly in terms of energy consumption and memory usage. A novel framework is proposed for detecting and mitigating HF attacks in resource-constrained RPL-based IoT networks. Through simulations using the Cooja simulator and Contiki 3.0 operating system, we demonstrate that HF attacks can significantly increase energy consumption by over 50% and deplete memory within minutes. The proposed framework leverages machine learning (ML) models trained on simulated network data to identify malicious traffic patterns accurately. The ML trained models were able to do HF attack detection with great accuracy. By denying malicious packets predicted by the model, the framework effectively mitigates the impact of HF attacks, conserving valuable resources and enhancing the resilience of RPL-based IoT networks.