Recent Advances in Graph Neural Networks with Optimized Attention and Long-Range CNN for Traffic Prediction and Resource Allocation in 6G Wireless Systems: A Systematic Review

Abstract

The evolution of 6G wireless systems has significantly increased the demand for intelligent traffic prediction and efficient resource allocation mechanisms to support ultra-reliable and low-latency communications. Traditional machine learning models fail to capture the complex spatial-temporal dependencies inherent in large-scale wireless and vehicular networks. Recently, Graph Neural Networks (GNNs), combined with optimized attention mechanisms and long-range Convolutional Neural Networks (CNNs), have emerged as powerful tools for modelling dynamic traffic patterns and optimizing resource allocation in 6G environments. This paper presents a systematic review of recent advances in GNN-based approaches for traffic prediction and resource management, focusing on studies published recently. The review highlights the effectiveness of spatial-temporal graph learning, attention-based feature extraction, and hybrid deep learning frameworks in improving prediction accuracy and network efficiency. Furthermore, it discusses the integration of reinforcement learning with GNNs for adaptive decision-making in dynamic wireless systems. Key challenges such as scalability, computational complexity, and real-time deployment are also examined. This study provides a comprehensive comparative analysis of existing models and identifies future research directions for developing intelligent and energy-efficient 6G wireless systems.