A Hybrid Intrusion Detection Framework For Energy-Constrained Iot Devices
Main Article Content
Abstract
The widespread adoption of Internet of Things (IoT) devices has enabled transformative applications across industries, yet it has also introduced significant security vulnerabilities, especially in energy-constrained environments. Conventional intrusion detection systems (IDS) often impose high computational and energy overheads, limiting their applicability in IoT networks. This paper presents a hybrid intrusion detection framework specifically designed for energy-limited IoT devices, combining signature-based detection, anomaly detection, and lightweight machine learning techniques. The framework ensures effective detection without sacrificing device longevity by utilizing cloud-based resources for advanced threat intelligence and edge computing for real-time local analysis. Comparing the suggested framework to current IDS techniques, experimental assessments on benchmark IoT datasets show that it delivers better detection accuracy, lower false positive rates, and improved energy consumption. The findings show that a hybrid, energy-aware IDS can successfully protect IoT networks from a variety of cyberthreats, providing a scalable and useful solution for both consumer and industrial IoT deployments.