Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG

Zhanghao Pan; Yanxiang Jiang; Yige Hang; Fu Chun Zheng

doi:10.1109/VTC2025-Fall65116.2025.11310087

Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG

Zhanghao Pan
, Yanxiang Jiang^*
, Yige Hang
, Fu Chun Zheng^*

^*Corresponding author for this work

Southeast University, Nanjing
Harbin Institute of Technology Shenzhen

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

This paper introduces a two-stage operational framework for computation offloading in the mobile edge computing (MEC)-enabled cell free massive multiple-input multiple-output (CF-mMIMO) systems. The computation offloading task is divided into data transmission and task computing sub-tasks to achieve fine-grained optimization of energy efficiency and reduce task complexity. A staged multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed to solve the energy optimization problem. In the transmission stage, user equipments (UEs) act as agents to optimize the allocation of transmission power, while in the computing stage, access point (AP) clusters act as agents to optimize the allocation of computational resources. The simulation results demonstrate that compared with the baseline algorithm, the proposed scheme can achieve a significant improvement in energy consumption during the transmission stage, and near-optimal performance with lower complexity during the computing stage.

Original language	English
Title of host publication	2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9798331503208
DOIs	https://doi.org/10.1109/VTC2025-Fall65116.2025.11310087
State	Published - 2025
Externally published	Yes
Event	2025 IEEE 102nd Vehicular Technology Conference, VTC 2025 - Chengdu, China Duration: 19 Oct 2025 → 22 Oct 2025

Publication series

Name	IEEE Vehicular Technology Conference
ISSN (Print)	1090-3038

Conference

Conference	2025 IEEE 102nd Vehicular Technology Conference, VTC 2025
Country/Territory	China
City	Chengdu
Period	19/10/25 → 22/10/25

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

SDG 7 Affordable and Clean Energy

Keywords

cell-free massive MIMO
mobile edge computing
multi-agent reinforcement learning

Access to Document

10.1109/VTC2025-Fall65116.2025.11310087

Cite this

Pan, Z., Jiang, Y., Hang, Y., & Zheng, F. C. (2025). Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG. In 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings (IEEE Vehicular Technology Conference). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/VTC2025-Fall65116.2025.11310087

@inproceedings{e8c76275564b4c32a4b26a5ff06619e9,

title = "Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG",

abstract = "This paper introduces a two-stage operational framework for computation offloading in the mobile edge computing (MEC)-enabled cell free massive multiple-input multiple-output (CF-mMIMO) systems. The computation offloading task is divided into data transmission and task computing sub-tasks to achieve fine-grained optimization of energy efficiency and reduce task complexity. A staged multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed to solve the energy optimization problem. In the transmission stage, user equipments (UEs) act as agents to optimize the allocation of transmission power, while in the computing stage, access point (AP) clusters act as agents to optimize the allocation of computational resources. The simulation results demonstrate that compared with the baseline algorithm, the proposed scheme can achieve a significant improvement in energy consumption during the transmission stage, and near-optimal performance with lower complexity during the computing stage.",

keywords = "cell-free massive MIMO, mobile edge computing, multi-agent reinforcement learning",

author = "Zhanghao Pan and Yanxiang Jiang and Yige Hang and Zheng, \{Fu Chun\}",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025 ; Conference date: 19-10-2025 Through 22-10-2025",

year = "2025",

doi = "10.1109/VTC2025-Fall65116.2025.11310087",

language = "英语",

series = "IEEE Vehicular Technology Conference",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

booktitle = "2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings",

address = "美国",

}

Pan, Z, Jiang, Y, Hang, Y & Zheng, FC 2025, Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG. in 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings. IEEE Vehicular Technology Conference, Institute of Electrical and Electronics Engineers Inc., 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025, Chengdu, China, 19/10/25. https://doi.org/10.1109/VTC2025-Fall65116.2025.11310087

Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG. / Pan, Zhanghao; Jiang, Yanxiang; Hang, Yige et al.
2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings. Institute of Electrical and Electronics Engineers Inc., 2025. (IEEE Vehicular Technology Conference).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG

AU - Pan, Zhanghao

AU - Jiang, Yanxiang

AU - Hang, Yige

AU - Zheng, Fu Chun

PY - 2025

Y1 - 2025

N2 - This paper introduces a two-stage operational framework for computation offloading in the mobile edge computing (MEC)-enabled cell free massive multiple-input multiple-output (CF-mMIMO) systems. The computation offloading task is divided into data transmission and task computing sub-tasks to achieve fine-grained optimization of energy efficiency and reduce task complexity. A staged multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed to solve the energy optimization problem. In the transmission stage, user equipments (UEs) act as agents to optimize the allocation of transmission power, while in the computing stage, access point (AP) clusters act as agents to optimize the allocation of computational resources. The simulation results demonstrate that compared with the baseline algorithm, the proposed scheme can achieve a significant improvement in energy consumption during the transmission stage, and near-optimal performance with lower complexity during the computing stage.

AB - This paper introduces a two-stage operational framework for computation offloading in the mobile edge computing (MEC)-enabled cell free massive multiple-input multiple-output (CF-mMIMO) systems. The computation offloading task is divided into data transmission and task computing sub-tasks to achieve fine-grained optimization of energy efficiency and reduce task complexity. A staged multi-agent deep deterministic policy gradient (MADDPG) algorithm is proposed to solve the energy optimization problem. In the transmission stage, user equipments (UEs) act as agents to optimize the allocation of transmission power, while in the computing stage, access point (AP) clusters act as agents to optimize the allocation of computational resources. The simulation results demonstrate that compared with the baseline algorithm, the proposed scheme can achieve a significant improvement in energy consumption during the transmission stage, and near-optimal performance with lower complexity during the computing stage.

KW - cell-free massive MIMO

KW - mobile edge computing

KW - multi-agent reinforcement learning

UR - https://www.scopus.com/pages/publications/105032447491

U2 - 10.1109/VTC2025-Fall65116.2025.11310087

DO - 10.1109/VTC2025-Fall65116.2025.11310087

M3 - 会议稿件

AN - SCOPUS:105032447491

T3 - IEEE Vehicular Technology Conference

BT - 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025-Fall - Proceedings

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2025 IEEE 102nd Vehicular Technology Conference, VTC 2025

Y2 - 19 October 2025 through 22 October 2025

ER -

Energy-Optimized Computation Offloading in MEC-Enabled Cell-Free Massive MIMO Systems with MADDPG

Abstract

Publication series

Conference

UN SDGs

Keywords

Access to Document

Other files and links

Fingerprint

Cite this