GCN-Based Throughput-Oriented Handover Management in Dense 5G Vehicular Networks

Authors:

Abstract
The rapid advancement of 5G has transformed vehicular networks, offering high bandwidth, low latency, and fast data rates essential for real-time applications in smart cities and vehicles. These improvements enhance traffic safety and entertainment services. However, 5G's limited coverage and frequent handovers, causing network instability from the “pingpong effect,” pose challenges in high-mobility environments. This paper presents TH-GCN (Throughput-oriented Graph Convolutional Network), a novel approach for optimizing handover management in dense 5G networks. Utilizing graph neural networks (GNNs), TH-GCN models vehicles and base stations as nodes in a dynamic graph with enriched features like signal quality, throughput, vehicle speed, and base station load. Integrating both user equipment and base station perspectives, this dualcentric approach enables adaptive, real-time handover decisions that improve stability. Simulations show that TH-GCN reduces handovers by up to 78% and improves signal quality by 10%, outperforming existing methods and positioning it as a key advancement in 5G vehicular networks.