Citation: Ortiz-Gomez, F.G.; Lei, L.;
Lagunas, E.; Martinez, R.; Tarchi, D.;
Querol, J.; Salas-Natera, M.A.;
Chatzinotas, S. Machine Learning for
Radio Resource Management in
Multibeam GEO Satellite Systems.
Electronics 2022, 11, 992. https://
doi.org/10.3390/electronics11070992
Academic Editor: Mauro Tropea
Received: 28 January 2022
Accepted: 16 March 2022
Published: 23 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Machine Learning for Radio Resource Management in
Multibeam GEO Satellite Systems
Flor G. Ortiz-Gomez
1,
* , Lei Lei
2
, Eva Lagunas
1
, Ramon Martinez
3
, Daniele Tarchi
4
, Jorge Querol
1
,
Miguel A. Salas-Natera
3
and Symeon Chatzinotas
1
1
Interdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg,
4365 Luxembourg, Luxembourg; eva.lagunas@uni.lu (E.L.); jorge.querol@uni.lu (J.Q.);
symeon.chatzinotas@uni.lu (S.C.)
2
School of Information and Communications Engineering, Xi’an Jiaotong University, Xi’an 710049, China;
lei.lei@xjtu.edu.cn
3
Information Processing and Telecommunications Center, Universidad Politecnica de Madrid,
28040 Madrid, Spain; ramon.martinez@upm.es (R.M.); miguel.salas@upm.es (M.A.S.-N.)
4
Department of Electrical,Electronic and Information Engineering, University of Bologna, 40136 Bologna, Italy;
daniele.tarchi@unibo.it
* Correspondence: flor.otiz@uni.lu
Abstract:
Satellite communications (SatComs) systems are facing a massive increase in traffic demand.
However, this increase is not uniform across the service area due to the uneven distribution of
users and changes in traffic demand diurnal. This problem is addressed by using flexible payload
architectures, which allow payload resources to be flexibly allocated to meet the traffic demand of
each beam. While optimization-based radio resource management (RRM) has shown significant
performance gains, its intense computational complexity limits its practical implementation in real
systems. In this paper, we discuss the architecture, implementation and applications of Machine
Learning (ML) for resource management in multibeam GEO satellite systems. We mainly focus on
two systems, one with power, bandwidth, and/or beamwidth flexibility, and the second with time
flexibility, i.e., beam hopping. We analyze and compare different ML techniques that have been
proposed for these architectures, emphasizing the use of Supervised Learning (SL) and Reinforcement
Learning (RL). To this end, we define whether training should be conducted online or offline based on
the characteristics and requirements of each proposed ML technique and discuss the most appropriate
system architecture and the advantages and disadvantages of each approach.
Keywords:
satellite communications;radio resource management; flexible payload; beam hopping;
machine learning; supervised learning; reinforcement learning
1. Introduction
One of the main challenges in designing future satellite broadband systems is how to
increase satellite revenues while meeting uneven and dynamic traffic demands [
1
,
2
]. In
this regard, a flexible payload is a promising solution to meet changing traffic demand
patterns. As a consequence, recent research interests have focused on designing a new
generation of flexible satellite payloads that enable radio resource management (RRM)
based on non-uniform traffic demand [
3
–
6
]. In that sense, Cocco et al. [
5
] represent the
problem of RRM for multibeam satellite as an objective function that minimizes the error
between the capacity offered and the capacity required. Nevertheless, a thorough analysis
of both the design of the payload architecture and resource management is required.
Kawamoto et al. [
7
] suggest that optimization techniques are a valid and efficient
approach to address the resource allocation problem. However, at a larger scale, the number
of resources to be managed, the constraints arising from the system and the massive
number of traffic demand situations typically may result in a problem that conventional
Electronics 2022, 11, 992. https://doi.org/10.3390/electronics11070992 https://www.mdpi.com/journal/electronics
