Power Optimization in Heterogeneous Cellular Networks Using Deep Reinforcement Learning Based Sleep Control Algorithm
International Journal of Engineering & Extended Technologies Research · 28.03.2026 · 08 (02)
DOI: 10.15662/ijeetr.2026.0802089 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Engineering & Extended Technologies Research, online erschienen am 28.03.2026. Verfasst von P. Deepa und 3 weiteren Autor:innen.
Abstract (Originalsprache)
The rapid growth of mobile users and increasing data traffic has significantly increased energy consumption in modern cellular networks. Heterogeneous Cellular Networks (HetNets), which consist of macro and small cells, are widely used to improve network coverage and capacity. However, the dense deployment of base stations leads to higher power consumption. In many cases, several base stations remain active even during low traffic conditions, resulting in unnecessary energy usage. Therefore, improving energy efficiency while maintaining network performance has become an important research challenge. This project proposes a Deep Reinforcement Learning (DRL) based sleep control algorithm to improve energy efficiency in heterogeneous cellular networks. The DRL agent continuously observes network parameters such as traffic load, number of active users, and base station utilization. Based on these observations, the agent intelligently decides whether a small cell base station should remain active or switch to sleep mode. During low traffic conditions, some small cells are dynamically switched to sleep mode to reduce energy consumption. When the traffic demand increases, the base stations can be reactivated to maintain service quality and network coverage. The proposed approach aims to reduce overall network power consumption while ensuring Quality of Service (QoS) and maintaining system performance. The effectiveness of the proposed DRL-based method is evaluated and compared with traditional algorithms such as Branch and Bound (BnB) and Dynamic methods. The results demonstrate that the DRL-based sleep control algorithm provides better energy efficiency and improved network performance in heterogeneous cellular networks
Zitieren
APA
Deepa, P., Sivaram, S., Surendran, B., & Tamilarasan, R. (2026). Power Optimization in Heterogeneous Cellular Networks Using Deep Reinforcement Learning Based Sleep Control Algorithm. International Journal of Engineering & Extended Technologies Research, 08(02). https://doi.org/10.15662/ijeetr.2026.0802089
Vancouver
Deepa P, Sivaram S, Surendran B, Tamilarasan R. Power Optimization in Heterogeneous Cellular Networks Using Deep Reinforcement Learning Based Sleep Control Algorithm. International Journal of Engineering & Extended Technologies Research. 2026;08(02). doi:10.15662/ijeetr.2026.0802089
BibTeX
@article{Deepa2026,
author = {Deepa, P. and Sivaram, S. and Surendran, B. and Tamilarasan, R.},
title = {Power Optimization in Heterogeneous Cellular Networks Using Deep Reinforcement Learning Based Sleep Control Algorithm},
journal = {International Journal of Engineering & Extended Technologies Research},
year = {2026},
volume = {08},
number = {02},
doi = {10.15662/ijeetr.2026.0802089}
}
EUVIMED