Artificial Intelligence-assisted Software Defined Networking for Securing ITS in Smart Cities

Rapid urbanization has caused cities to expand and grow rapidly with evolving technologies. This has introduced the concept of smart cities. The intelligent transportation system (ITS) is an important paradigm for any smart city. ITS is based on sensor deployment for data collection and controllers for data transmission. It increases the transportation network’s efficiency, safety, manageability, and reliability. Usually, traditional routing systems (TRS) are utilized for data exchange, making the network complicated and unreliable. Hence, software-defined networking (SDN)-based routing systems resolve issues encountered by the TRS. We have used machine learning (ML) models to distinguish between malicious and non-malicious data and transmit it accurately. We implemented various ML classification models such as decision tree (DT), random forest (RF), support vector machine (SVM), and XGBoost. Through implementation, we tried to increase data reliability and network safety. The performance of these models is compared using parameters like F1 score, accuracy, precision, recall, precision-recall (PR) curve, receiver operating characteristic (ROC) curve, and confusion matrix. Incorporating SDN-based systems in ITS can optimize the routing system’s performance and make the transportation network simple and reliable.