Hyun-Kyo Lim, Ju-Bong Kim, Youn-Hee Han,

Journal of Advanced Technology Research, Vol. 6, No. 2, pp. 10-15, Dec. 2021
10.11111/JATR.2021.6.2.010, Full Text:
Keywords: Virtual Network Embedding, Hierarchical Reinforcement Learning, Multi-Agent-based Reinforcement Learning
Abstract

Network virtualization logically divides physical network resources into one software-based network on each hardware equipment through virtualization to meet the requirements of each service. However, in order to efficiently provide network resources to virtual networks generated to meet the requirements of each service, virtual network embedding algorithms that can be embedded into real physical networks are needed. In this paper, we propose a virtual network embedding algorithm that applies to 1) hierarchical reinforcement learning for efficient exploration, and 2) multi-agent-based reinforcement learning for improve the performance of algorithms through collaboration of multiple agents. To verify the proposed algorithm, we prove its validity by performing comparisons with the latest reinforcement learning-based algorithms.

Statistics
Cite this article