Fusion-based Diversified Model for Internet of Vehicles: Leveraging Artificial Intelligence in Cloud Computing

Abstract

The Internet of Vehicles (IoV) is a distributed system that enables data connectivity between vehicles and vehicular ad hoc networks, ensuring efficient and secure information exchange with infrastructures. Challenges in IoV include security clustering related to packet loss during data exchange, real-time analysis of public communication, and the need for autonomous-vehicle technology development using machine learning (ML). ML-assisted IoV has made significant progress in communication with public networks and interaction with the immediate surroundings. This study presents an experimental foundation for the advancement of the IoV system. While support vector machine (SVM) offers a robust and accurate approach for clustering velocity and solving classification challenges related to security, it is primarily a binary classifier and faces limitations in handling multi-class classification. To address this, an artificial neural network (ANN) is proposed for effective packet loss management in the autonomous system, improving the physical layer's secure network and offering better packet loss experience using the Global Positioning System. The fusion-based diversified model not only enables IoV systems to compete with rivals but also provides key advantages to ensure consistent profitability in cloud-enabled IoV. This paradigm integrates cloud computing (CC) with in-vehicle networks and the Internet of Things, offering safety and infotainment applications for road users. Data collection and experiments are conducted using Network Simulator 2 to automate AI configuration in the IoV fusion system.