A machine learning approach for energy consumption in 5G networks

Energy consumption modeling in 5G networks is a complex task due to the variability in network configurations, traffic conditions, and the deployment of energy-saving techniques. Machine Learning (ML) offers a promising approach to address these challenges by leveraging data-driven insights for accurate predictions. This study aims to explore this idea by developing an ML-based model to predict energy consumption in 5G networks. We employ XGBoost, CatBoost, and Artificial Neural Networks (ANN), combined through a weighted average approach, to enhance prediction accuracy. Our findings indicate that the ensemble model significantly improves the estimation of energy consumption patterns, providing valuable insights for efficient energy management in 5G networks.