A deep reinforcement learning based research for optimal offloading decision
Currently, a concern about power resource constraints in the distribution environment is being voiced increasingly, where the increase of power consumption devices overwhelms the terminal load unaffordable and the quality of power consumption cannot be guaranteed. How to acquire the optimal offloadi...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
AIP Publishing LLC
2023-08-01
|
| Series: | AIP Advances |
| Online Access: | http://dx.doi.org/10.1063/5.0157491 |
| _version_ | 1797689658845429760 |
|---|---|
| author | Jianji Ren Donghao Yang Yongliang Yuan Huihui Wei Zhenxi Wang |
| author_facet | Jianji Ren Donghao Yang Yongliang Yuan Huihui Wei Zhenxi Wang |
| author_sort | Jianji Ren |
| collection | DOAJ |
| description | Currently, a concern about power resource constraints in the distribution environment is being voiced increasingly, where the increase of power consumption devices overwhelms the terminal load unaffordable and the quality of power consumption cannot be guaranteed. How to acquire the optimal offloading decision of power resources has become a problem that needs to be addressed urgently. To tackle this challenge, a novel reinforcement learning algorithm named Deep Q Network with a partial offloading strategy (DQNP) is proposed to optimize power resource allocation for high computational demands. In the DQNP, a coupled coordination degree model and Lyapunov algorithm are introduced, which trade-offs and decouples the relationships between local-edge and latency–energy consumption. To derive the optimal offloading decision, the resource computation utility function is selected as the objective function. In addition, model pruning is availed to further improve the training time and inference results. Results show that the proposed offloading mechanism can significantly decrease the function value and decline the weighted sum of latency and energy consumption by an average of 3.61%–7.31% relative to other state-of-the-art algorithms. Additionally, the energy loss in the power distribution process is successfully mitigated; furthermore, the effectiveness of the proposed algorithm is also verified. |
| first_indexed | 2024-03-12T01:48:30Z |
| format | Article |
| id | doaj.art-98854fac853e4e35aeac2831d7430623 |
| institution | Directory Open Access Journal |
| issn | 2158-3226 |
| language | English |
| last_indexed | 2024-03-12T01:48:30Z |
| publishDate | 2023-08-01 |
| publisher | AIP Publishing LLC |
| record_format | Article |
| series | AIP Advances |
| spelling | doaj.art-98854fac853e4e35aeac2831d74306232023-09-08T16:03:30ZengAIP Publishing LLCAIP Advances2158-32262023-08-01138085204085204-910.1063/5.0157491A deep reinforcement learning based research for optimal offloading decisionJianji Ren0Donghao Yang1Yongliang Yuan2Huihui Wei3Zhenxi Wang4School of Software, Henan Polytechnic University, Jiaozuo 454003, Henan, ChinaSchool of Software, Henan Polytechnic University, Jiaozuo 454003, Henan, ChinaSchool of Mechanical and Power Engineering, Henan Polytechnic University, Jiaozuo 454003, Henan, ChinaSchool of Software, Henan Polytechnic University, Jiaozuo 454003, Henan, ChinaSchool of Software, Henan Polytechnic University, Jiaozuo 454003, Henan, ChinaCurrently, a concern about power resource constraints in the distribution environment is being voiced increasingly, where the increase of power consumption devices overwhelms the terminal load unaffordable and the quality of power consumption cannot be guaranteed. How to acquire the optimal offloading decision of power resources has become a problem that needs to be addressed urgently. To tackle this challenge, a novel reinforcement learning algorithm named Deep Q Network with a partial offloading strategy (DQNP) is proposed to optimize power resource allocation for high computational demands. In the DQNP, a coupled coordination degree model and Lyapunov algorithm are introduced, which trade-offs and decouples the relationships between local-edge and latency–energy consumption. To derive the optimal offloading decision, the resource computation utility function is selected as the objective function. In addition, model pruning is availed to further improve the training time and inference results. Results show that the proposed offloading mechanism can significantly decrease the function value and decline the weighted sum of latency and energy consumption by an average of 3.61%–7.31% relative to other state-of-the-art algorithms. Additionally, the energy loss in the power distribution process is successfully mitigated; furthermore, the effectiveness of the proposed algorithm is also verified.http://dx.doi.org/10.1063/5.0157491 |
| spellingShingle | Jianji Ren Donghao Yang Yongliang Yuan Huihui Wei Zhenxi Wang A deep reinforcement learning based research for optimal offloading decision AIP Advances |
| title | A deep reinforcement learning based research for optimal offloading decision |
| title_full | A deep reinforcement learning based research for optimal offloading decision |
| title_fullStr | A deep reinforcement learning based research for optimal offloading decision |
| title_full_unstemmed | A deep reinforcement learning based research for optimal offloading decision |
| title_short | A deep reinforcement learning based research for optimal offloading decision |
| title_sort | deep reinforcement learning based research for optimal offloading decision |
| url | http://dx.doi.org/10.1063/5.0157491 |
| work_keys_str_mv | AT jianjiren adeepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT donghaoyang adeepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT yongliangyuan adeepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT huihuiwei adeepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT zhenxiwang adeepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT jianjiren deepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT donghaoyang deepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT yongliangyuan deepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT huihuiwei deepreinforcementlearningbasedresearchforoptimaloffloadingdecision AT zhenxiwang deepreinforcementlearningbasedresearchforoptimaloffloadingdecision |