DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs
The rapid growth of Automotive-Industry 5.0 and its emergence with beyond fifth-generation (B5G) communications, is making vehicular edge computing networks (VECNs) increasingly complex. The latency constraints of modern automotive applications make it difficult to run complex applications on vehicl...
Main Authors: | , , , , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Elsevier
2023-06-01
|
Series: | Journal of King Saud University: Computer and Information Sciences |
Subjects: | |
Online Access: | http://www.sciencedirect.com/science/article/pii/S1319157823000423 |
_version_ | 1797796104392146944 |
---|---|
author | Muhammad Ayzed Mirza Junsheng Yu Salman Raza Moez Krichen Manzoor Ahmed Wali Ullah Khan Khaled Rabie Thokozani Shongwe |
author_facet | Muhammad Ayzed Mirza Junsheng Yu Salman Raza Moez Krichen Manzoor Ahmed Wali Ullah Khan Khaled Rabie Thokozani Shongwe |
author_sort | Muhammad Ayzed Mirza |
collection | DOAJ |
description | The rapid growth of Automotive-Industry 5.0 and its emergence with beyond fifth-generation (B5G) communications, is making vehicular edge computing networks (VECNs) increasingly complex. The latency constraints of modern automotive applications make it difficult to run complex applications on vehicle on-board units (OBUs). While multi-access edge computing (MEC) can facilitate task offloading to execute these applications, it is still a challenge to access them promptly and optimally. Traditional algorithms struggle to guarantee accuracy in such dynamic environment, but deep reinforcement learning (DRL) methods offer improved accuracy, robustness, and real-time decision-making capabilities. In this paper, we propose a DRL-based mobility, contact, and load aware cooperative task offloading (DCTO) scheme. DCTO is designed for both cellular and mmWave radio access technologies (RATs), and both binary and partial offloading mechanisms. DCTO targets delay minimization by opportunistically switching RATs and offloading mechanisms. We consider relative efficacy and neutrality factors as key performance indicators and use them to derive the DRL agent’s reward function. Extensive evaluations demonstrate that the DCTO scheme exhibits a substantial enhancement in task success rate, with an increase from 2.61% to 21.34%. It also improves the efficacy factor from 1.38 to 3.52 and reduces the neutrality factor from 4.99 to 0.76. Furthermore, the average task processing time is reduced by a range of 3.77% to 24.15%. Additionally, the DCTO scheme outperforms the other evaluated schemes in terms of reward and TFPS ratio. |
first_indexed | 2024-03-13T03:28:09Z |
format | Article |
id | doaj.art-35ea90d49ce24f4d918c75b28bfd860b |
institution | Directory Open Access Journal |
issn | 1319-1578 |
language | English |
last_indexed | 2024-03-13T03:28:09Z |
publishDate | 2023-06-01 |
publisher | Elsevier |
record_format | Article |
series | Journal of King Saud University: Computer and Information Sciences |
spelling | doaj.art-35ea90d49ce24f4d918c75b28bfd860b2023-06-25T04:42:39ZengElsevierJournal of King Saud University: Computer and Information Sciences1319-15782023-06-01356101512DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNsMuhammad Ayzed Mirza0Junsheng Yu1Salman Raza2Moez Krichen3Manzoor Ahmed4Wali Ullah Khan5Khaled Rabie6Thokozani Shongwe7BUPT-QMUL EM Theory and Application International Research Lab, Beijing University of Posts and Telecommunications, Beijing 100876, ChinaBUPT-QMUL EM Theory and Application International Research Lab, Beijing University of Posts and Telecommunications, Beijing 100876, China; School of Physics and Electronic Information, Anhui Normal University, Wuhu 241003, China; School of Intelligence and Digital Engineering, Luoyang Vocational College of Science and Technology, Luoyang 471000, China; Corresponding authors.Department of Computer Science, National Textile University, Faisalabad 37610, PakistanFaculty of CSIT, Al-Baha University, Saudi Arabia ReDCAD Laboratory, University of Sfax, TunisiaSchool of Computer and Information Science and also with Institute for AI Industrial Technology Research, Hubei Engineering University, Xiaogan 432000, ChinaInterdisciplinary Centre for Security, Reliability and Trust (SnT), University of Luxembourg, Luxembourg City 1855, LuxembourgDepartment of Engineering, Manchester Metropolitan University, Manchester M15 6BH, UK; Department of Electrical and Electronic Engineering Technology, University of Johannesburg, Auckland Park, Johannesburg 2006, South Africa; Corresponding authors.Department of Electrical and Electronic Engineering Technology, University of Johannesburg, Auckland Park, Johannesburg 2006, South AfricaThe rapid growth of Automotive-Industry 5.0 and its emergence with beyond fifth-generation (B5G) communications, is making vehicular edge computing networks (VECNs) increasingly complex. The latency constraints of modern automotive applications make it difficult to run complex applications on vehicle on-board units (OBUs). While multi-access edge computing (MEC) can facilitate task offloading to execute these applications, it is still a challenge to access them promptly and optimally. Traditional algorithms struggle to guarantee accuracy in such dynamic environment, but deep reinforcement learning (DRL) methods offer improved accuracy, robustness, and real-time decision-making capabilities. In this paper, we propose a DRL-based mobility, contact, and load aware cooperative task offloading (DCTO) scheme. DCTO is designed for both cellular and mmWave radio access technologies (RATs), and both binary and partial offloading mechanisms. DCTO targets delay minimization by opportunistically switching RATs and offloading mechanisms. We consider relative efficacy and neutrality factors as key performance indicators and use them to derive the DRL agent’s reward function. Extensive evaluations demonstrate that the DCTO scheme exhibits a substantial enhancement in task success rate, with an increase from 2.61% to 21.34%. It also improves the efficacy factor from 1.38 to 3.52 and reduces the neutrality factor from 4.99 to 0.76. Furthermore, the average task processing time is reduced by a range of 3.77% to 24.15%. Additionally, the DCTO scheme outperforms the other evaluated schemes in terms of reward and TFPS ratio.http://www.sciencedirect.com/science/article/pii/S1319157823000423Automotive-Industry 5.0Vehicular Edge Computing (VEC)Task offloadingBeyond fifth-generation (B5G)Deep Reinforcement Learning (DRL) |
spellingShingle | Muhammad Ayzed Mirza Junsheng Yu Salman Raza Moez Krichen Manzoor Ahmed Wali Ullah Khan Khaled Rabie Thokozani Shongwe DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs Journal of King Saud University: Computer and Information Sciences Automotive-Industry 5.0 Vehicular Edge Computing (VEC) Task offloading Beyond fifth-generation (B5G) Deep Reinforcement Learning (DRL) |
title | DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs |
title_full | DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs |
title_fullStr | DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs |
title_full_unstemmed | DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs |
title_short | DRL-assisted delay optimized task offloading in automotive-industry 5.0 based VECNs |
title_sort | drl assisted delay optimized task offloading in automotive industry 5 0 based vecns |
topic | Automotive-Industry 5.0 Vehicular Edge Computing (VEC) Task offloading Beyond fifth-generation (B5G) Deep Reinforcement Learning (DRL) |
url | http://www.sciencedirect.com/science/article/pii/S1319157823000423 |
work_keys_str_mv | AT muhammadayzedmirza drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT junshengyu drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT salmanraza drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT moezkrichen drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT manzoorahmed drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT waliullahkhan drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT khaledrabie drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns AT thokozanishongwe drlassisteddelayoptimizedtaskoffloadinginautomotiveindustry50basedvecns |