Abstract
Internet of Things (IoT) is one of the significant emerging domains in computing. IoT structures the base of future infrastructures, developing the advancement of future smart cities that are naturally moderate. However, IoT is vulnerable to different cyber-attacks. The most critical issues of the IoT are security and privacy. In this research, the integration of IoT with cloud and fog computing can present an improved platform to support IoT smart city applications. Cloud computing can provide data storage at top-level management and fog computing can offer various services for supporting smart city applications based on IoT. The integrated architecture of the IoT-Fog-Cloud computing model is presented in this research. For the analysis based on Network-based Intrusion Detection System (NIDS) anomaly detection, the Improved Naive Bayes (INB) classifier based on Principal Component Analysis (PCA) technique is proposed in this work. The UNSW-NB15 dataset is used for the evaluation of the attack detection model. The PCA technique is used to extract features of the dataset, and the INB classifier is used for the classification of attacks. This technique is proposed to enhance the efficiency of the anomaly detection and to improve the performance analysis in terms of accuracy, precision, recall, and detection rate. The INB-PCA achieved 92.48% accuracy and 95.35% detection rate.