DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs

<p dir="ltr">Coverage holes pose critical challenges to reliability of wireless networks and their quality of service (QoS) and therefore should be avoided in the coverage design. In this paper, we address this issue through the deployment of unmanned aerial vehicles (UAVs) as mobile...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Bahareh Jafari (22501715) (author)
مؤلفون آخرون: Mazen Hasna (16904661) (author), Hossein Pishro-Nik (22501718) (author), Nizar Zorba (16888728) (author), Tamer Khattab (16870086) (author), Hamid Saeedi (21399992) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513523995901952
author Bahareh Jafari (22501715)
author2 Mazen Hasna (16904661)
Hossein Pishro-Nik (22501718)
Nizar Zorba (16888728)
Tamer Khattab (16870086)
Hamid Saeedi (21399992)
author2_role author
author
author
author
author
author_facet Bahareh Jafari (22501715)
Mazen Hasna (16904661)
Hossein Pishro-Nik (22501718)
Nizar Zorba (16888728)
Tamer Khattab (16870086)
Hamid Saeedi (21399992)
author_role author
dc.creator.none.fl_str_mv Bahareh Jafari (22501715)
Mazen Hasna (16904661)
Hossein Pishro-Nik (22501718)
Nizar Zorba (16888728)
Tamer Khattab (16870086)
Hamid Saeedi (21399992)
dc.date.none.fl_str_mv 2025-04-28T03:00:00Z
dc.identifier.none.fl_str_mv 10.1109/ojcoms.2025.3564837
dc.relation.none.fl_str_mv https://figshare.com/articles/journal_contribution/DRL-Based_UAV_Path_Planning_for_Coverage_Hole_Avoidance_Energy_Consumption_and_Outage_Time_Minimization_Trade-Offs/30845393
dc.rights.none.fl_str_mv CC BY 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Engineering
Communications engineering
Control engineering, mechatronics and robotics
Information and computing sciences
Distributed computing and systems software
Machine learning
Coverage hole
QoS
Path planning
UAV
Mechanical energy
Deep reinforcement learning (DRL)
dc.title.none.fl_str_mv DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
dc.type.none.fl_str_mv Text
Journal contribution
info:eu-repo/semantics/publishedVersion
text
contribution to journal
description <p dir="ltr">Coverage holes pose critical challenges to reliability of wireless networks and their quality of service (QoS) and therefore should be avoided in the coverage design. In this paper, we address this issue through the deployment of unmanned aerial vehicles (UAVs) as mobile base stations, and we propose specific UAV path planning. A point is said to be in a coverage hole if the coverage probability for that point is below a certain threshold, e.g., 90%. This definition is more suitable for applications such as surveillance or sensor networks. In this paper, we target applications such as wireless communications for which QoS requirement allow only for short time disconnections, i.e., minimal outage time. As such, in addition to avoiding coverage holes, we should also make the outage time as small as possible. By deploying a deep reinforcement learning algorithm, we find optimal UAV paths based on the two families of trajectories: spiral and oval curves, to tackle different design considerations and constraints, in terms of QoS, energy consumption and coverage hole avoidance. We show that for a typical point on the cell, there is a trade-off between minimizing the maximum outage time length and consumed mechanical energy. Our observations indicate that such a trade-off is more pronounced for spiral trajectories compared to oval trajectories, but both of them are useful depending on the QoS and energy constraints imposed by the system.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Communications Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" rel="noreferrer" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcoms.2025.3564837" target="_blank">https://dx.doi.org/10.1109/ojcoms.2025.3564837</a></p>
eu_rights_str_mv openAccess
id Manara2_d0539ccfaabfe7670fef19dc3bd06230
identifier_str_mv 10.1109/ojcoms.2025.3564837
network_acronym_str Manara2
network_name_str Manara2
oai_identifier_str oai:figshare.com:article/30845393
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY 4.0
spelling DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-OffsBahareh Jafari (22501715)Mazen Hasna (16904661)Hossein Pishro-Nik (22501718)Nizar Zorba (16888728)Tamer Khattab (16870086)Hamid Saeedi (21399992)EngineeringCommunications engineeringControl engineering, mechatronics and roboticsInformation and computing sciencesDistributed computing and systems softwareMachine learningCoverage holeQoSPath planningUAVMechanical energyDeep reinforcement learning (DRL)<p dir="ltr">Coverage holes pose critical challenges to reliability of wireless networks and their quality of service (QoS) and therefore should be avoided in the coverage design. In this paper, we address this issue through the deployment of unmanned aerial vehicles (UAVs) as mobile base stations, and we propose specific UAV path planning. A point is said to be in a coverage hole if the coverage probability for that point is below a certain threshold, e.g., 90%. This definition is more suitable for applications such as surveillance or sensor networks. In this paper, we target applications such as wireless communications for which QoS requirement allow only for short time disconnections, i.e., minimal outage time. As such, in addition to avoiding coverage holes, we should also make the outage time as small as possible. By deploying a deep reinforcement learning algorithm, we find optimal UAV paths based on the two families of trajectories: spiral and oval curves, to tackle different design considerations and constraints, in terms of QoS, energy consumption and coverage hole avoidance. We show that for a typical point on the cell, there is a trade-off between minimizing the maximum outage time length and consumed mechanical energy. Our observations indicate that such a trade-off is more pronounced for spiral trajectories compared to oval trajectories, but both of them are useful depending on the QoS and energy constraints imposed by the system.</p><h2 dir="ltr">Other Information</h2><p dir="ltr">Published in: IEEE Open Journal of the Communications Society<br>License: <a href="https://creativecommons.org/licenses/by/4.0/" rel="noreferrer" target="_blank">https://creativecommons.org/licenses/by/4.0/</a><br>See article on publisher's website: <a href="https://dx.doi.org/10.1109/ojcoms.2025.3564837" target="_blank">https://dx.doi.org/10.1109/ojcoms.2025.3564837</a></p>2025-04-28T03:00:00ZTextJournal contributioninfo:eu-repo/semantics/publishedVersiontextcontribution to journal10.1109/ojcoms.2025.3564837https://figshare.com/articles/journal_contribution/DRL-Based_UAV_Path_Planning_for_Coverage_Hole_Avoidance_Energy_Consumption_and_Outage_Time_Minimization_Trade-Offs/30845393CC BY 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/308453932025-04-28T03:00:00Z
spellingShingle DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
Bahareh Jafari (22501715)
Engineering
Communications engineering
Control engineering, mechatronics and robotics
Information and computing sciences
Distributed computing and systems software
Machine learning
Coverage hole
QoS
Path planning
UAV
Mechanical energy
Deep reinforcement learning (DRL)
status_str publishedVersion
title DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
title_full DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
title_fullStr DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
title_full_unstemmed DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
title_short DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
title_sort DRL-Based UAV Path Planning for Coverage Hole Avoidance: Energy Consumption and Outage Time Minimization Trade-Offs
topic Engineering
Communications engineering
Control engineering, mechatronics and robotics
Information and computing sciences
Distributed computing and systems software
Machine learning
Coverage hole
QoS
Path planning
UAV
Mechanical energy
Deep reinforcement learning (DRL)