Ensemble-based cyber intrusion detection for robust smart city protection

The rapid rise of 5G networks has accelerated the integration of smart cities, marking the emergence of increased intelligence in urban environments, often referred to as Smart Cities. This swift integration has interconnected a wide range of devices and systems, thereby exposing them to potential vulnerabilities. As a result, a smart urban landscape has emerged where valuable and sensitive information is shared without adequate attention to security considerations. Given these challenges, it is essential to implement an effective cloud-based Intrusion Detection System (IDS) for the security of smart cities. This work examines the reliability and robustness of various ensemble learning models, focusing on evaluating the performance and efficiency of an IDS strategy based on machine learning to enhance the security of IoT in smart urban networks. We conducted experimental procedures on three commonly used datasets to achieve the objectives of our study. The results obtained from these procedures are crucial for developing practical IDS solutions that address the ever-changing challenges posed by diverse, smart, cloud-based network traffic systems in smart cities.