Design of an Intelligent Energy Management System for BTS Sites in Telecommunication Networks

Document Type : Research Article

Authors

1 Faculty of Science, Mahallat Institute of Higher Education, Mahallat, 37811-51958, Iran

2 Faculty of Engineering, Mahallat Institute of Higher Education, Mahallat, 37811-51958, Iran

10.22091/jaem.2026.16422.1049

Abstract

Energy consumption in telecommunication Base Transceiver Station (BTS) sites has become a major challenge for network operators due to the increasing demand for mobile services and the growing operational costs of communication infrastructures. This study proposes an intelligent energy management framework that integrates deep learning and reinforcement learning techniques to optimize energy consumption in BTS sites while maintaining network Quality of Service (QoS). An LSTM-based model is employed to predict future energy consumption and traffic load patterns, whereas a Proximal Policy Optimization (PPO) reinforcement learning agent is utilized to adaptively control the cooling system and minimize unnecessary energy usage. The proposed framework was evaluated through a seven-day pilot deployment at an urban BTS site equipped with 4G radio equipment and an air-conditioning-based cooling system. Experimental results demonstrated an average energy saving of 15.5% compared with a conventional control strategy, without any observed QoS degradation. The highest energy saving reached 21.0% under high-traffic operating conditions. Furthermore, the LSTM model achieved mean relative prediction errors below 4% for urban sites and below 3% for suburban and remote sites, indicating reliable short-term forecasting performance. Analysis of the pilot data showed that the proposed controller maintained effective operation under varying traffic loads and environmental conditions, demonstrating its adaptability to dynamic network scenarios. The results highlight the potential of combining deep learning and reinforcement learning approaches for the development of intelligent, energy-efficient, and sustainable telecommunication infrastructures.

Keywords

Main Subjects

Advances in Energy and Materials Research

Articles in Press, Accepted Manuscript
Available Online from 28 September 2025

Files

Share

How to cite

Statistics