Lightweight Explainable Edge AI Framework for IoT Network Anomaly Detection

This work presents a lightweight explainable Edge AI framework for IoT network anomaly detection. The proposed model is designed to improve real-time detection performance on resource-constrained edge devices. It integrates machine learning techniques with explainable AI methods such as SHAP and LIME to enhance interpretability and trust. Experimental evaluation shows improved accuracy, reduced latency, and better efficiency compared to traditional machine learning models such as SVM and LSTM. The framework is suitable for deployment in real-world IoT environments requiring low-latency and high-security monitoring.