Intelligent Relay Selection in 5G D2D Communication: Leveraging Machine Learning for Enhanced Coverage

In the evolving landscape of 5G networks, Device-to-Device (D2D) communication has emerged as a significant technology to offload traffic, enhance user experience, and expand network coverage. While D2D promises seamless connectivity, efficient relay selection remains a challenge, particularly in dynamic communication environments. This paper introduces an intelligent relay selection mechanism that leverages machine learning, specifically Artificial Neural Networks (ANN) with Radial Basis Function Neural Network (RBFNN) approach, to optimize D2D communication in 5G networks. By integrating a threshold-based relay selection and combining with the predictive capabilities of ANN, we aim to improve overall network coverage. Our method dynamically adjusts selection criteria based on real-time network conditions, ensuring optimal relay selection and minimizing communication breakdowns. Initial simulation results reveal that our approach exceeds traditional techniques, showcasing significant improvements in the coverage area, data output, and reduced inactivity. This research shows the way for a more adaptive, intelligent and efficient D2D communication framework in 5G systems.