Enhancing Cybersecurity in Smart Building Sensor Networks through AI-Driven Intrusion Detection

Abstract The Internet of Things (IoT) is increasingly integrated into smart buildings to enhance automation, efficiency, and occupant experience. However, these distributed sensor networks introduce significant cybersecurity risks. This paper presents an Artificial Intelligence–based Intrusion Detection System (AI-IDS) that leverages supervised machine learning algorithms, including Support Vector Machine (SVM), Random Forest (RF), K-Nearest Neighbor (KNN), and Decision Tree (DT). The system is trained on two benchmark datasets, CICIDS2017 and IoT-23, with preprocessing techniques such as Synthetic Minority Oversampling Technique (SMOTE), Principal Component Analysis (PCA), and Recursive Feature Elimination (RFE) applied to improve performance. Experimental results demonstrate that the SVM model achieved the highest detection accuracy (99.02%) and offered the best balance between accuracy and training time. These findings indicate that the proposed AI-IDS can provide efficient, real-time security for smart building environments, enhancing resilience against evolving cyber threats.