TY - GEN
T1 - Deep Reinforcement Learning for Backhaul Link Selection for Network Slices in IAB Networks
AU - Morgado, António J.
AU - Saghezchi, Firooz B.
AU - Fondo-Ferreiro, Pablo
AU - Gil-Castineira, Felipe
AU - Papaioannou, Maria
AU - Ramantas, Kostas
AU - Rodriguez, Jonathan
N1 - Funding Information:
This work was supported in part by Xunta de Galicia (Spain) under grant ED481B-2022-019 and by EXPLOR project funded by H2020-MSCA-RISE-2019 (grant agreement ID: 872897).
Publisher Copyright:
© 2023 IEEE.
PY - 2024/2/26
Y1 - 2024/2/26
N2 - Integrated Access and Backhaul (IAB) has been recently proposed by 3GPP to enable network operators to deploy fifth generation (5G) mobile networks with reduced costs. In this paper, we propose to use IAB to build a dynamic wireless backhaul network capable to provide additional capacity to those Base Stations (BS) experiencing congestion momentarily. As the mobile traffic demand varies across time and space, and the number of slice combinations deployed in a BS can be prohibitively high, we propose to use Deep Reinforcement Learning (DRL) to select, from a set of candidate BSs, the one that can provide backhaul capacity for each of the slices deployed in a congested BS. Our results show that a Double Deep Q-Network (DDQN) agent using a fully connected neural network and the Rectified Linear Unit (ReLU) activation function with only one hidden layer is capable to perform the BS selection task successfully, without any failure during the test phase, after being trained for around 20 episodes.
AB - Integrated Access and Backhaul (IAB) has been recently proposed by 3GPP to enable network operators to deploy fifth generation (5G) mobile networks with reduced costs. In this paper, we propose to use IAB to build a dynamic wireless backhaul network capable to provide additional capacity to those Base Stations (BS) experiencing congestion momentarily. As the mobile traffic demand varies across time and space, and the number of slice combinations deployed in a BS can be prohibitively high, we propose to use Deep Reinforcement Learning (DRL) to select, from a set of candidate BSs, the one that can provide backhaul capacity for each of the slices deployed in a congested BS. Our results show that a Double Deep Q-Network (DDQN) agent using a fully connected neural network and the Rectified Linear Unit (ReLU) activation function with only one hidden layer is capable to perform the BS selection task successfully, without any failure during the test phase, after being trained for around 20 episodes.
KW - backhaul link selection
KW - deep reinforcement learning
KW - integrated access and backhaul
KW - machine learning
KW - network slicing
KW - resource allocation
U2 - 10.1109/GLOBECOM54140.2023.10436900
DO - 10.1109/GLOBECOM54140.2023.10436900
M3 - Conference contribution
AN - SCOPUS:85187376326
T3 - Proceedings - IEEE Global Communications Conference, GLOBECOM
SP - 6267
EP - 6272
BT - GLOBECOM 2023 - 2023 IEEE Global Communications Conference
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2023 IEEE Global Communications Conference, GLOBECOM 2023
Y2 - 4 December 2023 through 8 December 2023
ER -