Analysis of Algorithms of Machine Learning Applied to the Assignment of Resources in the Grid of Access for 5G Networks with Segmentation

This study explores the use of supervised learning and Q-learning to optimize dynamic preamble allocation in a single-cell 5G network, supporting SDG 9 (Industry, Innovation, and Infrastructure) by improving infrastructure resilience. The network is divided into three service categories: mMTC, eMBB, and H2H communication, contributing to SDG 11 (Sustainable Cities and Communities). The algorithms adjust preamble allocation based on traffic fluctuations, aiming to minimize collisions. While both algorithms improved performance, Q-learning showed better adaptability, especially under high traffic, aligning with SDG 12 (Responsible Consumption and Production) by optimizing resource use and reducing energy consumption.