| Title | AI-Driven Intrusion Detection for Resource-Constrained Internet of Things Networks |
| Research Area | Internet of Things |
| Abstract | The rapid expansion of the Internet of Things (IoT) has connected billions of heterogeneous devices to communication networks and cloud-based services. Although IoT provides substantial benefits in automation, monitoring, healthcare, transportation, agriculture, and industrial operations, its large and distributed attack surface creates significant cybersecurity challenges. Conventional intrusion detection systems frequently struggle to operate effectively in IoT environments because IoT net-works contain resource-constrained devices, diverse communication protocols, irregular traffic pat-terns, and continuously changing attack behaviors. Artificial intelligence and machine learning pro-vide an opportunity to develop adaptive intrusion detection mechanisms capable of identifying suspi-cious network behavior from observed traffic characteristics. This paper presents an AI-driven intru-sion detection framework designed specifically for resource-constrained IoT environments. The pro-posed framework combines traffic preprocessing, feature reduction, lightweight machine learning, anomaly scoring, and edge-based decision making. Instead of requiring every IoT device to execute computationally expensive detection algorithms, the architecture transfers major analytical opera-tions to edge gateways. The proposed design also considers class imbalance, false-positive reduc-tion, model updating, explainability, and computational efficiency |
| Keywords | Internet of Things, Intrusion Detection, Artificial Intelligence, Machine Learning, Edge Computing, Anomaly Detection, Cybersecurity, IoT Networks, Lightweight Security |
| Paper Status | Published |
| Volume | 2 |
| Issue | 4 |
| Published On | 15/07/2026 |
| Published File |
IJSRTD_4887.pdf
|