| Title | AI-Driven Intrusion Detection in Internet of Things Networks: A Lightweight and Adaptive Framework for Next-Generation Cybersecurity |
| Research Area | Internet of Things |
| Abstract | This paper presents a comprehensive examination of AI-driven intrusion detection for IoT environ-ments and proposes a conceptual lightweight and adaptive framework. The proposed architecture dis-tributes security intelligence across IoT devices, network monitoring components, edge gateways, and centralized security infrastructure. Lightweight anomaly screening is performed near the data source, while computationally intensive analysis is reserved for suspicious or uncertain events. The framework further incorporates adaptive learning, explainable detection, privacy-aware data processing, and a feedback mechanism for continuous improvement. The paper analyzes major IoT threats, conventional IDS techniques, machine learning approaches, deep learning architectures, edge intelligence, federated learning, evaluation metrics, deployment limitations, and future research opportunities. Rather than optimizing only classification accuracy, the proposed framework considers detection effectiveness, latency, resource consumption, scalability, privacy, adaptability, and interpretability. The study pro-vides a foundation for future experimental research into practical AI-enabled intrusion detection sys-tems capable of protecting heterogeneous and evolving IoT environments. |
| Keywords | Internet of Things, Artificial Intelligence, Machine Learning, Intrusion Detection, Cyber-security, Deep Learning, Edge Computing, Anomaly Detection, Federated Learning, IoT Security |
| Paper Status | Published |
| Volume | 2 |
| Issue | 3 |
| Published On | 16/05/2026 |
| Published File |
IJSRTD_4879.pdf
|