Deep Learning Neural Networks-based Traffic Predictors for V2X Communication Networks Marina Magdy Saady1*, Hatim Ghazi Zaini2, Mohamed H. Essai3, Sahar A. El_Rahman4, Osama A. Omer1, Ali R. Abdellah3, and Shaima EL Nazer5

Marina Magdy^1*Hatim Ghazi Zaini²Mohamed H. Essai³Sahar A. El_Rahman⁴Osama A. Omer¹Ali R. Abdellah⁵Shaima EL Nazer⁶

• ¹Department of Electrical Engineering, Faculty of Engineering, Aswan University, Aswan 81542, Egypt, Aswan, Egypt
• ²Computer Engineering Department, College of Computer and Information Technology, Taif University, Taif 21944, Saudi Arabia, Taif, Saudi Arabia
• ³Department of Electrical Engineering, Faculty of Engineering, Al-Azhar University, Qena 83513, Egypt, Qena, Egypt
• ⁴Department of Computer Systems program-Electrical Engineering, Faculty of Engineering-Shoubra, Benha University, Cairo, Egypt, Cairo,, Egypt
• ⁵Department of Electrical Engineering, Faculty of Engineering, Al-Azhar University, Qena 83513, Aswan, Egypt
• ⁶Nile Higher Institute of Engineering and Technology, Mansoura 35516, Egypt, Mansoura, Egypt

Vehicle-to-everything (V2X) communication is a promising technology for enhancing road safety, traffic efficiency, and the availability of infotainment services in 5G networks and beyond networks. However, the effective sharing of traffic information remains a significant challenge. To address this, AI-based systems offer potential solutions. By predicting traffic patterns on dense networks, these systems can improve traffic management, mitigate congestion, increase network safety and reliability, and improve energy efficiency. This research investigates the application of Recurrent Neural Networks (RNNs) and Convolutional Neural Networks (CNNs) for accurate and efficient V2X traffic prediction. We explored the impact of various hyperparameters, including loss functions and optimizers, on the performance of these models. Our findings indicate that Gated Recurrent Unit (GRU) models, particularly with the Mean Squared Error (MSE) loss function and Adam optimizer, consistently outperform Long Short-Term Memory (LSTM) and Bidirectional Long Short-Term Memory (BiLSTM) models in terms of both accuracy and computational efficiency. For CNN models, the Rectified Linear Unit (ReLU) activation function, coupled with the Adam optimizer, demonstrated superior performance in terms of Root Mean Square Error (RMSE) and computational complexity. By comparing our results with existing literature, we highlight the advantages of our proposed models in terms of accuracy, efficiency, and robustness.