Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning

A low earth orbit constellation beam hopping resource scheduling method based on multi-agent deep reinforcement learning was proposed to meet the requirements of low earth orbit constellation beam hopping resource scheduling. The mapping relationship between the satellite and the service area was es...

Full description

Saved in:
Bibliographic Details
Main Authors: ZHANG Chen, XU Yangwei, LI Wanjing, WANG Wei, ZHANG Gengxin
Format: Article
Language:zho
Published: Editorial Department of Journal on Communications 2025-01-01
Series:Tongxin xuebao
Subjects:
Online Access:http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2025009/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850194631351861248
author ZHANG Chen
XU Yangwei
LI Wanjing
WANG Wei
ZHANG Gengxin
author_facet ZHANG Chen
XU Yangwei
LI Wanjing
WANG Wei
ZHANG Gengxin
author_sort ZHANG Chen
collection DOAJ
description A low earth orbit constellation beam hopping resource scheduling method based on multi-agent deep reinforcement learning was proposed to meet the requirements of low earth orbit constellation beam hopping resource scheduling. The mapping relationship between the satellite and the service area was established by optimizing the access of multi-target satellite selection. On this basis, according to the diversity of service types and QoS requirements, based on the concept of mixture of experts, a resource scheduling multi-agent was constructed to carry out real-time decision scheduling of on-board resources and beam hopping patterns. The simulation results show that compared with the traditional methods, the proposed resource scheduling method can not only meet the performance requirements of different services on delay and throughput, but also effectively balance the algorithm complexity. At the same time, the algorithm can adapt to the converged transmission requirements of diversified services, cope with the uneven spatiotemporal distribution and dynamic changes of traffic and have strong generalization ability.
format Article
id doaj-art-af3cc67d314a4dd5bbe21b18a44eff5d
institution OA Journals
issn 1000-436X
language zho
publishDate 2025-01-01
publisher Editorial Department of Journal on Communications
record_format Article
series Tongxin xuebao
spelling doaj-art-af3cc67d314a4dd5bbe21b18a44eff5d2025-08-20T02:13:56ZzhoEditorial Department of Journal on CommunicationsTongxin xuebao1000-436X2025-01-0146355182296236Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learningZHANG ChenXU YangweiLI WanjingWANG WeiZHANG GengxinA low earth orbit constellation beam hopping resource scheduling method based on multi-agent deep reinforcement learning was proposed to meet the requirements of low earth orbit constellation beam hopping resource scheduling. The mapping relationship between the satellite and the service area was established by optimizing the access of multi-target satellite selection. On this basis, according to the diversity of service types and QoS requirements, based on the concept of mixture of experts, a resource scheduling multi-agent was constructed to carry out real-time decision scheduling of on-board resources and beam hopping patterns. The simulation results show that compared with the traditional methods, the proposed resource scheduling method can not only meet the performance requirements of different services on delay and throughput, but also effectively balance the algorithm complexity. At the same time, the algorithm can adapt to the converged transmission requirements of diversified services, cope with the uneven spatiotemporal distribution and dynamic changes of traffic and have strong generalization ability.http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2025009/low earth orbit satellitebeam hoppingdeep reinforcement learningresource scheduling
spellingShingle ZHANG Chen
XU Yangwei
LI Wanjing
WANG Wei
ZHANG Gengxin
Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
Tongxin xuebao
low earth orbit satellite
beam hopping
deep reinforcement learning
resource scheduling
title Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
title_full Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
title_fullStr Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
title_full_unstemmed Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
title_short Research on low earth orbit constellation beam hopping resource scheduling based on multi-agent deep reinforcement learning
title_sort research on low earth orbit constellation beam hopping resource scheduling based on multi agent deep reinforcement learning
topic low earth orbit satellite
beam hopping
deep reinforcement learning
resource scheduling
url http://www.joconline.com.cn/zh/article/doi/10.11959/j.issn.1000-436x.2025009/
work_keys_str_mv AT zhangchen researchonlowearthorbitconstellationbeamhoppingresourceschedulingbasedonmultiagentdeepreinforcementlearning
AT xuyangwei researchonlowearthorbitconstellationbeamhoppingresourceschedulingbasedonmultiagentdeepreinforcementlearning
AT liwanjing researchonlowearthorbitconstellationbeamhoppingresourceschedulingbasedonmultiagentdeepreinforcementlearning
AT wangwei researchonlowearthorbitconstellationbeamhoppingresourceschedulingbasedonmultiagentdeepreinforcementlearning
AT zhanggengxin researchonlowearthorbitconstellationbeamhoppingresourceschedulingbasedonmultiagentdeepreinforcementlearning