Enhanced Traffic Management in IoT-Integrated Internet of Vehicles (IoV) with Optimized Routing and Deep Learning Algorithm

The integration of the Internet of Things (IoT) into vehicular networks has paved the way for the exciting development of the Internet of Vehicles (IoV) within Intelligent Transportation Systems (ITS). These breakthroughs bring about enhancements in urban transportation services, including improved quality and interaction, reduced costs, more efficient resource utilization, and enhanced traffic management capabilities. The incorporation of IoV into transportation has the potential to address various traffic-related challenges in smart cities. Nevertheless, optimizing routing and monitoring network traffic presents significant challenges, primarily due to limited node-to-node communication capabilities. To tackle these problems, this study proposes the BiLSTM-GreedyNet, which combines the Improved Greedy Perimeter Stateless Routing Algorithm with the Bi-LSTM neural network technology in a 5G network environment. The system delivers impressive results, as evidenced by simulations in MATLAB, boasting a remarkable 98.22% accuracy rate, a low 34.2 MJ energy consumption, and a high Packet Delivery Ratio (PDR) of 91%, all tested with a network of 100 nodes.