Edge-Resident Intrusion Detection System (IDoS): Leveraging AI for Anomaly Detection in IoT Devices

Abstract

This research presents a novel way to improve the security of Internet of Things (IoT) devices using an Artificial Intelligence (AI)-powered Edge-Resident Intrusion Detection System (IDoS). The suggested approach addresses latency, bandwidth, and scalability issues by bringing anomaly detection capabilities closer to the source of IoT data through the use of edge computing. By comparing normal and aberrant patterns in real-time device behaviour, the Edge-Resident IDoS uses sophisticated machine learning algorithms. This technology optimises bandwidth and minimises reliance on centralised cloud solutions by being deployed directly on edge devices or local servers. It also reduces latency. A proactive approach to security is ensured by AI-driven anomaly detection, which sees possible attacks before they escalate. The system's effectiveness in promoting safe and effective IoT installations is highlighted by discussing the practical ramifications, which include effects on compliance, energy efficiency, and reliability. The Edge-Resident IoT ecosystem, which contributes to robust and responsive IoT ecosystems, is a revolutionary step towards safeguarding connected settings through the convergence of edge computing and AI.