Paper · 2025 · Open access

Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing

Liu Yang, YANG Yanyan, CHAI Zhengyi

DOAJ (DOAJ: Directory of Open Access Journals) · Published 1 January 2025

Open full textCite this paper

In brief

“Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing” is a 2025 paper by Liu Yang and 2 co-authors, published in DOAJ (DOAJ: Directory of Open Access Journals). OpenAlex counts 0 citations as of 19 November 2026. It is open access under a CC BY-SA licence: the full text is free to read at DOAJ (DOAJ: Directory of Open Access Journals).

Citations
0
Authors
3

Abstract

ObjectiveIn mobile edge computing, the edge clients can offload their computing tasks to edge nodes to reduce their own computing pressure and energy consumption. However, when the edge node receives a large number of computing tasks, it will cause its own high load, resulting in some computing tasks cannot be processed in time to meet the low latency requirements of edge clients, and in extreme cases some of the tasks will even be discarded. Most of the existing computation offloading methods struggle to achieve the desired optimum in environments where the load on edge nodes is dynamically uncertain. To address this problem, a distributed task offloading algorithm based on deep reinforcement learning (DRL) is proposed to minimize the long-term cost of the system so that each edge client can make independent offloading decisions without knowledge of the edge node loads and offloading decisions of other clients.MethodsA multi-user and multi node edge computing environment was considered a multi-user and multi node edge computing environment. Edge client processing models and edge node processing models were established. The optimization of the system includes two parts: delay and energy consumption during the task offloading process. In order to minimize the total cost of the system, the above optimization problem is modeled as a Markov decision process (MDP), and a distributed task offloading algorithm based on DRL is proposed. The offloading process is based on MDP and the architecture of state, action, reward function, and neural network were constructed. To ensure the estimation of the long-term cost, a combination of Dueling Deep Q-Network (Dueling DQN) and Double Deep Q-Network (Double DQN) techniques is introduced to improve the training efficiency and stability of the policy.Results and Discussions In the experiment, simulation experiments were conducted using the deep learning framework TensorFlow. a MEC system environment consisting of 30 edge clients and 3 edge servers is simulated, with edge servers evenly deployed throughout the entire system. In addition to the latency and energy consumption of edge clients, the performance of the algorithm is also evaluated by considering the power of successful task offloading as an indicator. The results show that our algorithm can effectively reduce the average energy consumption and latency of tasks and improve the success rate of offloading compared to the benchmark algorithm. Moreover, under the same experimental parameter settings, the proposed algorithm DMOO has a faster convergence speed and better optimization results for the problem.ConclusionsThe problem of distributed task offloading from multiple servers to multiple users in MEC systems was investigated, and a distributed offloading…

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Yang, L., Yanyan, Y., & Zhengyi, C. (2025). Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing. DOAJ (DOAJ: Directory of Open Access Journals). https://doaj.org/article/f600e85e21854deeb05f47291b3d419d

MLA 9

Yang, Liu, et al. “Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing.” DOAJ (DOAJ: Directory of Open Access Journals), 2025. https://doaj.org/article/f600e85e21854deeb05f47291b3d419d.

Chicago

Yang, Liu, YANG Yanyan, and CHAI Zhengyi. 2025. “Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing.” DOAJ (DOAJ: Directory of Open Access Journals). https://doaj.org/article/f600e85e21854deeb05f47291b3d419d.

Harvard

Yang, L., Yanyan, Y. and Zhengyi, C. (2025) ‘Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing’, DOAJ (DOAJ: Directory of Open Access Journals). Available at: https://doaj.org/article/f600e85e21854deeb05f47291b3d419d.

IEEE

L. Yang, Y. Yanyan, and C. Zhengyi, “Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing,” DOAJ (DOAJ: Directory of Open Access Journals), 2025.

BibTeX

@article{yang2025distributed,
  title     = {{Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing}},
  author    = {Yang, Liu and Yanyan, YANG and Zhengyi, CHAI},
  journal   = {DOAJ (DOAJ: Directory of Open Access Journals)},
  year      = {2025},
  url       = {https://doaj.org/article/f600e85e21854deeb05f47291b3d419d}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Topic and keywords

Physical Sciences › Computer Science › Computer Networks and Communications › IoT and Edge/Fog Computing

  • mobile edge computing
  • deep reinforcement learning
  • energy consumption
  • Markov decision process
  • dueling DQN
  • double deep Q-network
  • decentralized decision-making
  • reward function design

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Yang, L., Yanyan, Y., & Zhengyi, C. (2025). Distributed offloading algorithm based on deep reinforcement learning in mobile edge computing. DOAJ (DOAJ: Directory of Open Access Journals). https://doaj.org/article/f600e85e21854deeb05f47291b3d419d

How many times has this paper been cited?

0 times, according to OpenAlex on 19 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at DOAJ (DOAJ: Directory of Open Access Journals): sign in and open the full text from this page.

Who wrote this paper?

Liu Yang, YANG Yanyan, CHAI Zhengyi.

Bibliographic data from OpenAlex (CC0), last checked 19 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.