Learning-Based User Association and Cache Replacement for Communication Latency Reduction in Small Cell Network 


Vol. 45,  No. 6, pp. 1129-1136, Jun.  2020
10.7840/kics.2020.45.6.1129


PDF
  Abstract

In this paper, we consider a joint user association and cache replacement problem for reducing the latency caused by wireless transmission failures in a small cell network. We show our latency minimization problem can be formulated by Markov decision process (MDP) and propose a novel reinforcement learning algorithm to derive an effective policy for the problem. The proposed algorithm introduces permutation layers into the neural network of deep Q-network (DQN) algorithm to resolve the limitation of the conventional DQN algorithm by mitigating the correlation between adjacent actions. The simulation results validate that the reduced correlation with the proposed neural network design facilitates the learning in the right direction and brings performance improvement in terms of the communication latency.

  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.


  Cite this article

[IEEE Style]

J. Jung, S. Jeon, J. Hong, "Learning-Based User Association and Cache Replacement for Communication Latency Reduction in Small Cell Network," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 6, pp. 1129-1136, 2020. DOI: 10.7840/kics.2020.45.6.1129.

[ACM Style]

Jae-Wook Jung, Sang-Eun Jeon, and Jun-Pyo Hong. 2020. Learning-Based User Association and Cache Replacement for Communication Latency Reduction in Small Cell Network. The Journal of Korean Institute of Communications and Information Sciences, 45, 6, (2020), 1129-1136. DOI: 10.7840/kics.2020.45.6.1129.

[KICS Style]

Jae-Wook Jung, Sang-Eun Jeon, Jun-Pyo Hong, "Learning-Based User Association and Cache Replacement for Communication Latency Reduction in Small Cell Network," The Journal of Korean Institute of Communications and Information Sciences, vol. 45, no. 6, pp. 1129-1136, 6. 2020. (https://doi.org/10.7840/kics.2020.45.6.1129)