Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning

Abstract In highly dynamic vehicular networking scenarios, when Vehicle-to-Infrastructure links and Vehicle-to-Vehicle links share spectrum resources, the traditional distributed resource allocation method lacks global optimization and fails to respond to environmental changes in a timely manner, wh...

Full description

Saved in:
Bibliographic Details
Main Authors: Qingli Liu, Yongjie Ma
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-15982-x
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849226393146097664
author Qingli Liu
Yongjie Ma
author_facet Qingli Liu
Yongjie Ma
author_sort Qingli Liu
collection DOAJ
description Abstract In highly dynamic vehicular networking scenarios, when Vehicle-to-Infrastructure links and Vehicle-to-Vehicle links share spectrum resources, the traditional distributed resource allocation method lacks global optimization and fails to respond to environmental changes in a timely manner, which leads to low spectral efficiency of the system. A resource allocation method based on federated multi-agent deep reinforcement learning is proposed for Vehicular Networking communication, by fusing Asynchronous Federated Learning (AFL) and Multi-Agent Deep Deterministic Policy Gradient (MADDPG). Synergistic optimization of resource allocation. First, vehicles as agent dynamically optimize spectrum access, power control, and bandwidth allocation based on local channel states through the collaborative policy of MADDPG to reduce cross-link interference. Second, the asynchronous federation architecture is designed, where vehicles independently upload local model parameters to the global server, dynamically adjust the aggregation weights according to the real-time channel quality, and optimize the update of global model parameters. Finally, the global model parameters are fed back to the vehicles to further optimize the local resource allocation strategy, thus improving the system spectrum efficiency. The simulation results show that the system spectrum efficiency is improved by 19.1% on average compared with the centralized DDPG, MADDPG, MAPPO and FL-DuelingDQN algorithms in the Vehicle Networking scenario, while the transmission success rate of the V2V link is improved by 9.3% on average, and the total capacity of the V2I link is increased by 16.1% on average.
format Article
id doaj-art-f5c8a52597934be585b7591c7c0cc201
institution Kabale University
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-f5c8a52597934be585b7591c7c0cc2012025-08-24T11:25:23ZengNature PortfolioScientific Reports2045-23222025-08-0115111810.1038/s41598-025-15982-xCommunication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learningQingli Liu0Yongjie Ma1Key Laboratory of Communication and Network, Dalian UniversityKey Laboratory of Communication and Network, Dalian UniversityAbstract In highly dynamic vehicular networking scenarios, when Vehicle-to-Infrastructure links and Vehicle-to-Vehicle links share spectrum resources, the traditional distributed resource allocation method lacks global optimization and fails to respond to environmental changes in a timely manner, which leads to low spectral efficiency of the system. A resource allocation method based on federated multi-agent deep reinforcement learning is proposed for Vehicular Networking communication, by fusing Asynchronous Federated Learning (AFL) and Multi-Agent Deep Deterministic Policy Gradient (MADDPG). Synergistic optimization of resource allocation. First, vehicles as agent dynamically optimize spectrum access, power control, and bandwidth allocation based on local channel states through the collaborative policy of MADDPG to reduce cross-link interference. Second, the asynchronous federation architecture is designed, where vehicles independently upload local model parameters to the global server, dynamically adjust the aggregation weights according to the real-time channel quality, and optimize the update of global model parameters. Finally, the global model parameters are fed back to the vehicles to further optimize the local resource allocation strategy, thus improving the system spectrum efficiency. The simulation results show that the system spectrum efficiency is improved by 19.1% on average compared with the centralized DDPG, MADDPG, MAPPO and FL-DuelingDQN algorithms in the Vehicle Networking scenario, while the transmission success rate of the V2V link is improved by 9.3% on average, and the total capacity of the V2I link is increased by 16.1% on average.https://doi.org/10.1038/s41598-025-15982-xVehicular networksResource allocationDeep reinforcement learningAsynchronous federated learning
spellingShingle Qingli Liu
Yongjie Ma
Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
Scientific Reports
Vehicular networks
Resource allocation
Deep reinforcement learning
Asynchronous federated learning
title Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
title_full Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
title_fullStr Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
title_full_unstemmed Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
title_short Communication resource allocation method in vehicular networks based on federated multi-agent deep reinforcement learning
title_sort communication resource allocation method in vehicular networks based on federated multi agent deep reinforcement learning
topic Vehicular networks
Resource allocation
Deep reinforcement learning
Asynchronous federated learning
url https://doi.org/10.1038/s41598-025-15982-x
work_keys_str_mv AT qingliliu communicationresourceallocationmethodinvehicularnetworksbasedonfederatedmultiagentdeepreinforcementlearning
AT yongjiema communicationresourceallocationmethodinvehicularnetworksbasedonfederatedmultiagentdeepreinforcementlearning