Advanced Predictive Modeling for Enhancing Traffic Forecasting in Emerging Cellular Networks

The complexity and dynamicity of cellular emerging networks pose significant challenges to maintaining consistent quality of service (QoS). This research investigates the effectiveness of machine learning and deep learning algorithms in optimizing performance and enhancing user experience in 5G environments. We propose SARIMA, Prophet, LSTM and GRU models for adaptive resource allocation and accurate traffic prediction. The experimental evaluation demonstrates significant improvements in forecasting precision (MSE=1.28% using LSTM model) and good spatial traffic patterns using k-means clustering.