A4 Article in conference proceedings
Reinforcement Learning Based Mobility Load Balancing with the Cell Individual Offset (2021)

Asghari, M. Z., Ozturk, M., & Hämäläinen, J. (2021). Reinforcement Learning Based Mobility Load Balancing with the Cell Individual Offset. In Proceedings of the IEEE 93rd Vehicular Technology Conference (VTC2021-Spring). IEEE. IEEE Vehicular Technology Conference. https://doi.org/10.1109/VTC2021-Spring51267.2021.9448785

JYU authors or editors

Publication details

All authors or editors: Asghari, Muhammad Zeeshan; Ozturk, Metin; Hämäläinen, Jyri

Parent publication: Proceedings of the IEEE 93rd Vehicular Technology Conference (VTC2021-Spring)

Place and date of conference: Virtual Event, 25.-28.4.2021

eISBN: 978-1-7281-8964-2

Journal or series: IEEE Vehicular Technology Conference

ISSN: 1090-3038

eISSN: 2577-2465

Publication year: 2021

Publisher: IEEE

Publication country: United States

Publication language: English

DOI: https://doi.org/10.1109/VTC2021-Spring51267.2021.9448785

Publication open access: Not open

Publication channel open access:

Abstract

In this study, we focus on the cell individual offset (CIO) parameter in the handover process, which represents the willingness of a cell to admit the incoming handovers. However, it is challenging to tune the CIO parameter, as any poor implementation can lead to undesired outcomes, such as making the neighboring cells over-loaded while decreasing the traffic load of the cell. In this work, a reinforcement learning-based approach for parameter selection is introduced, since it is quite convenient for dynamically changing environments. In that regard, two different techniques, namely Q-learning and SARSA, are proposed, as they are known for their multi-objective optimization capabilities. Moreover, fixed CIO values are used as a benchmark for the proposed methods for comparison purposes. Results reveal that the reinforcement learning assisted mobility load balancing (MLB) approach can alleviate the burden on the overloaded cells while keeping the neighboring cells at some reasonable load levels. The proposed methods outperform the fixed-parameter solution in terms of the given metric.

Keywords: wireless networks; mobile communication networks; wireless data transmission; machine learning

Contributing organizations

Ministry reporting: Yes

Preliminary JUFO rating: 1