Simulation Design of Reinforcement-Based Migration System in Software-Defined Networking Using Q-Learning


Gana Raj, Avinash Imanuel and Tan, Saw Chin (2024) Simulation Design of Reinforcement-Based Migration System in Software-Defined Networking Using Q-Learning. Communications in Computer and Information Science, 1911. pp. 341-351. ISSN 1865-0929

Full text not available from this repository.


An innovative strategy, Software-Defined Networking (SDN), provides improved network programmability, flexibility, and scalability. However, a lot of recent material on the transition from traditional network designs to SDN has recommended providing static heuristic algorithms for determining the node migration sequence in legacy networks. The approach is deemed unfeasible for use in real-world circumstances and has issues managing the changing nature of network traffic. Reinforcement learning (RL) has therefore been proposed for use in SDN domains such flow entry management, controller placement, and routing selection. In the study, we first examine the relevant research on reinforcement learning applications on SDN conducted by previous researchers and identify the difficulties in SDN migration. Finally, using Q-learning in this context, we proposed a reinforcement learning technique to get around these problems and carefully transition a network’s legacy nodes to SDN nodes. Thus, we have discussed designs for applying reinforcement learning to a hSDN deployment in a legacy network in this article.

Item Type: Article
Uncontrolled Keywords: Software-defined networking
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television
Divisions: Faculty of Computing and Informatics (FCI)
Depositing User: Ms Nurul Iqtiani Ahmad
Date Deposited: 31 Jan 2024 02:34
Last Modified: 31 Jan 2024 02:34


Downloads per month over past year

View ItemEdit (login required)