Adaptive Resource Allocation in Emerging High-mobility Networks Using Hybrid Deep Learning Models

Resource allocation in 5G and Beyond 5G (B5G) networks is critical for maintaining high performance and Quality of Service (QoS), especially in high-mobility environments. In this paper, we present a novel, mobility-aware resource allocation framework utilizing a hybrid Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) model to predict downlink and uplink bitrates accurately. By capturing both spatial and temporal dependencies in network conditions and usermobility patterns, the hybrid CNN-LSTM model dynamically adjusts resource allocation to enhance network efficiency and user experience. The proposed model is evaluated on a comprehensive dataset incorporating various signal quality indicators, user speed, and network traffic. Experimental results demonstrate that the hybrid model outperforms traditional approaches, achieving lower Mean Squared Error (MSE) and higher R-squared values for bitrate predictions. This study provides valuable insights into the effectiveness of deep learning for adaptive resource allocation in next-generation B5G networks, enabling more reliable service under dynamic mobility conditions.