Distributed Deep Reinforcement Learning-Based Energy Efficiency Maximization in 3D Cellular Networks 


Vol. 48,  No. 8, pp. 942-949, Aug.  2023
10.7840/kics.2023.48.8.942


PDF
  Abstract

In this paper, we consider the multiple unmanned aerial vehicle-base station(UBS)-based 3D cellular networks to provide air-to-ground(A2G) communication coverage to moving ground users. Especially, to alleviate the short network lifetime problem of the UBS networks, we aim to control the movement and the transmission power of UBS so that maximizing the network-wide energy efficiency. However, considering the dynamic environment in which ground users move, deriving the optimal solution to the problem is significantly difficult with existing iterative methods or optimization methods. Therefore, in this paper, we propose a distributed deep Q-network(DQN)-based UBS control method. Also, to show the advantages of the distributed learning, we introduce two centralized learning methods, and then we consider the two centralized learning method, multi-agent distributed Q-learning(MD-QL) and greedy action(GA) methods as benchmarks. Conclusionally, we verify that the performance of the proposed method outperforms the conventional methods according to the movement speed of the ground user and the number of UBSs.

  Statistics
Cumulative Counts from November, 2022
Multiple requests among the same browser session are counted as one view. If you mouse over a chart, the values of data points will be shown.


  Related Articles
  Cite this article

[IEEE Style]

S. Lee, T. Ban, H. Lee, "Distributed Deep Reinforcement Learning-Based Energy Efficiency Maximization in 3D Cellular Networks," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 8, pp. 942-949, 2023. DOI: 10.7840/kics.2023.48.8.942.

[ACM Style]

Seungmin Lee, Tae-Won Ban, and Howon Lee. 2023. Distributed Deep Reinforcement Learning-Based Energy Efficiency Maximization in 3D Cellular Networks. The Journal of Korean Institute of Communications and Information Sciences, 48, 8, (2023), 942-949. DOI: 10.7840/kics.2023.48.8.942.

[KICS Style]

Seungmin Lee, Tae-Won Ban, Howon Lee, "Distributed Deep Reinforcement Learning-Based Energy Efficiency Maximization in 3D Cellular Networks," The Journal of Korean Institute of Communications and Information Sciences, vol. 48, no. 8, pp. 942-949, 8. 2023. (https://doi.org/10.7840/kics.2023.48.8.942)
Vol. 48, No. 8 Index

Search
(IN TITLE, AUTHOR, ABSTRACT,KEYWORDS)

Advanced Search ▼

POPULAR KEYWORDS
(TOP 10 KEYWORDS)

Recent Publications
(LAST 3 YEARS)






pISSN : 1226 - 4717
eISSN : 2287 - 3880