基于深度扩散确定性策略梯度的Wi-Fi网络性能优化

刘铁; 方旭明; 何蓉

doi:10.11959/j.issn.2096-3750.2025.00494

您当前的位置：

首页 >

文章列表页 >

基于深度扩散确定性策略梯度的Wi-Fi网络性能优化

理论与技术 | 更新时间：2026-01-04

- 基于深度扩散确定性策略梯度的Wi-Fi网络性能优化
- Deep diffusion deterministic policy gradient based performance optimization for Wi-Fi network
- 物联网学报 2025年9卷第4期页码：105-112
- 作者机构：
  
  西南交通大学信息科学与技术学院，四川成都 611756
- 作者简介：
  
  [ "刘铁（1999‒ ），男，西南交通大学信息科学与技术学院硕士生，主要研究方向为Wi-Fi网络性能优化、Wi-Fi标准。" ]
  [ "方旭明（1962‒ ），男，博士，西南交通大学信息科学与技术学院教授，主要研究方向为通信感知计算一体化网络、Wi-Fi 网络、智能交通移动通信系统。" ]
  [ "何蓉（1974‒ ），女，博士，西南交通大学信息科学与技术学院副教授，主要研究方向为Wi-Fi网络、无线资源分配、通信感知计算一体化网络。" ]
- 基金信息：
  
  国家自然科学基金资助项目(62071393);四川省重点研发计划项目(2024YFHZ0093)
- DOI：10.11959/j.issn.2096-3750.2025.00494
  中图分类号：
- 收稿：2025-02-19，
  
  修回：2025-06-10，
  
  纸质出版：2025-12-10
- 稿件说明：
移动端阅览
刘铁,方旭明,何蓉.基于深度扩散确定性策略梯度的Wi-Fi网络性能优化[J].物联网学报,2025,09(04):105-112.

LIU Tie,FANG Xuming,HE Rong.Deep diffusion deterministic policy gradient based performance optimization for Wi-Fi network[J].Chinese Journal on Internet of Things,2025,09(04):105-112.
刘铁,方旭明,何蓉.基于深度扩散确定性策略梯度的Wi-Fi网络性能优化[J].物联网学报,2025,09(04):105-112. DOI： 10.11959/j.issn.2096-3750.2025.00494.

LIU Tie,FANG Xuming,HE Rong.Deep diffusion deterministic policy gradient based performance optimization for Wi-Fi network[J].Chinese Journal on Internet of Things,2025,09(04):105-112. DOI： 10.11959/j.issn.2096-3750.2025.00494.

摘要

Wi-Fi网络性能优化通常是多参数、多目标动态优化问题，在数学建模上面临巨大挑战。深度强化学习（DRL

deep reinforcement learning）不需要复杂的数学建模，近年来被广泛应用于Wi-Fi网络性能优化。同时，生成扩散模型（GDMs

generative diffusion models）在多个领域中对复杂数据分布的建模取得了显著进展。因此，将DRL与GDMs相结合可以增强其对Wi-Fi网络性能优化的能力。Wi-Fi网络的典型介质访问控制（MAC

medium access control）接入机制是分布式协调功能（DCF

distributed coordination function），在竞争终端数量较多时，其性能会显著下降。提出了一种深度扩散确定性策略梯度（D3PG

deep diffusion deterministic policy gradient）算法，将扩散模型与深度确定性策略梯度（DDPG

deep deterministic policy gradient）框架相结合，以优化Wi-Fi网络性能。此外，还提出了一种基于D3PG算法的接入机制，联合调整竞争窗口和聚合帧长度。仿真实验表明，该机制在密集Wi-Fi场景中显著优于现有Wi-Fi标准的MAC接入机制，在竞争用户数量急剧增加时，仍能保持吞吐量性能稳定。

Abstract

The optimization of Wi-Fi network performance typically constitutes a multi-parameter

multi-objective dynamic optimization problem

which presents significant challenges in mathematical modeling. Deep reinforcement learning (DRL)

which obviates the need for complex mathematical formulations

has been widely applied in recent years to optimize Wi-Fi network performance. Meanwhile

generative diffusion models (GDMs) have achieved remarkable progress in modeling complex data distributions across various domains. Therefore

integrating DRL with GDMs can further enhance its capabilities in optimizing Wi-Fi network performance. The typical medium access control (MAC) mechanism in Wi-Fi network is the distributed coordination function (DCF)

whose performance significantly degrades as the number of contending terminals increases. A deep diffusion deterministic policy gradient (D3PG) algorithm was proposed

which integrated diffusion models with the deep deterministic policy gradient (DDPG) framework to optimize Wi-Fi network performance. In addition

an access mechanism that jointly adjusted the contention window and the aggregation frame length based on the D3PG algorithm was proposed. Simulations have demonstrated that this mechanism significantly outperforms existing Wi-Fi standards in dense Wi-Fi scenarios

maintaining throughput performance even as the number of users increases sharply.

关键词

Keywords

references

DENG C L , FANG X M , HAN X , et al . IEEE 802.11be Wi-Fi 7: new challenges and opportunities [J ] . IEEE Communications Surveys Tutorials , 2020 , 22 ( 4 ): 2136 - 2166 .

MOZAFFARIAHRAR E , THEOLEYRE F , MENTH M . A survey of Wi-Fi 6: technologies, advances, and challenges [J ] . Future Internet , 2022 , 14 ( 10 ): 293 .

LOPEZ-PEREZ D , GARCIA-RODRIGUEZ A , GALATI-GIORDANO L , et al . IEEE 802.11be extremely high throughput: the next generation of Wi-Fi technology beyond 802.11ax [J ] . IEEE Communications Magazine , 2019 , 57 ( 9 ): 113 - 119 .

WANG J , HAO Y L , YANG C . The current progress and future prospects of path loss model for terrestrial radio propagation [J ] . Electronics , 2023 , 12 ( 24 ): 4959 .

VASUDEVAN M , YUKSEL M . Machine learning for radio propagation modeling: a comprehensive survey [J ] . IEEE Open Journal of the Communications Society , 2024 , 5 : 5123 - 5153 .

伏玉笋 , 杨根科 . 人工智能在移动通信中的应用: 挑战与实践 [J ] . 通信学报 , 2020 , 41 ( 9 ): 190 - 201 .

FU Y S , YANG G K . Application of artificial intelligence in mobile communication: challenge and practice [J ] . Journal on Communications , 2020 , 41 ( 9 ): 190 - 201 .

KAUR A , KUMAR K . A comprehensive survey on machine learning approaches for dynamic spectrum access in cognitive radio networks [J ] . Journal of Experimental Theoretical Artificial Intelligence , 2022 , 34 ( 1 ): 1 - 40 .

SAI S , GARG A , JHAWAR K , et al . A comprehensive survey on artificial intelligence for unmanned aerial vehicles [J ] . IEEE Open Journal of Vehicular Technology , 2023 , 4 : 713 - 738 .

QUEIRÓS R , ALMEIDA E N , FONTES H , et al . Wi-Fi rate adaptation using a simple deep reinforcement learning approach [C ] // Proceedings of the 2022 IEEE Symposium on Computers and Communications (ISCC) . Piscataway : IEEE Press , 2022 : 1 - 3 .

CROITORU F A , HONDRU V , IONESCU R T , et al . Diffusion models in vision: a survey [J ] . IEEE Transactions on Pattern Analysis and Machine Intelligence , 2023 , 45 ( 9 ): 10850 - 10869 .

GOZALO-BRIZUELA R , GARRIDO-MERCHAN E C . ChatGPT is not all you need . A state of the art review of large generative AI models[EB ] . 2023

EPSTEIN Z , HERTZMANN A , AKTEN M , et al . Art and the science of generative AI [J ] . Science , 2023 , 380 ( 6650 ): 1110 - 1111 .

JO A . The promise and peril of generative AI [J ] . Nature , 2023 , 614 ( 1 ): 214 - 216 .

URKUDE P , CHARDE D , PIDURKAR S , et al . Multi-domain content generation using advanced machine learning techniques [C ] // Proceedings of the 2024 ICT for Intelligent Systems . Singapore : Springer , 2024 : 541 - 551 .

TRABUCCO B , DOHERTY K , GURINAS M , et al . Effective data augmentation with diffusion models [EB ] . 2023 .

DU H Y , ZHANG R C , LIU Y Q , et al . Enhancing deep reinforcement learning: a tutorial on generative diffusion models in network optimization [J ] . IEEE Communications Surveys Tutorials , 2024 , 26 ( 4 ): 2611 - 2646 .

DU H Y , LI Z H , NIYATO D , et al . Diffusion-based reinforcement learning for edge-enabled AI-generated content services [J ] . IEEE Transactions on Mobile Computing , 2024 , 23 ( 9 ): 8902 - 8918 .

KHAN M A A , BEEN K , BIPLOB B . A study on delay, throughput and traffic measurement for Wi-Fi connected stations based on MAC sublayer [J ] . IEEE Communications Surveys Tutorials , 2019 , 20 ( 1 ): 197 - 216 .

MANZOOR S , YIN Y C , GAO Y Y , et al . A systematic study of IEEE 802.11 DCF network optimization from theory to testbed [J ] . IEEE Access , 2020 , 8 : 154114 - 154132 .

LI Y , LV Z W , FAN Z H , et al . Adaptive two-step binary exponential backoff strategy for random access [C ] // Proceedings of the 2023 IEEE 6th Information Technology, Networking, Electronic and Automation Control Conference (ITNEC) . Piscataway : IEEE Press , 2023 : 1104 - 1109 .

UMEHARA D , YUAN J H . Success prioritized binary exponential backoff algorithm for non-persistent CSMA [C ] // Proceedings of the 2019 IEEE Global Communications Conference (GLOBECOM) . Piscataway : IEEE Press , 2020 : 1 - 6 .

BIANCHI G . Performance analysis of the IEEE 802.11 distributed coordination function [J ] . IEEE Journal on Selected Areas in Communications , 2000 , 18 ( 3 ): 535 - 547 .

SUZUKI K , YAMAZAKI S . Throughput maximization based on optimized frame-aggregation levels for IEEE 802.11 WLANs [J ] . IEEE Communications Letters , 2021 , 25 ( 5 ): 1725 - 1728 .

EL H B N , BUSSON A , RIVANO H . Analytical study of frame aggregation level to infer IEEE 802.11 network load [C ] // Proceedings of the 2020 International Wireless Communications and Mobile Computing (IWCMC) . Piscataway : IEEE Press , 2020 : 952 - 957 .

CAMPS-MUR D , GOMONY M D , PÉREZ-COSTA X , et al . Leveraging 802.11n frame aggregation to enhance QoS and power consumption in Wi-Fi networks [J ] . Computer Networks , 2012 , 56 ( 12 ): 2896 - 2911 .

CORONADO E , THOMAS A , RIGGIO R . Adaptive ML-based frame length optimisation in enterprise SD-WLANs [J ] . Journal of Network and Systems Management , 2020 , 28 ( 4 ): 850 - 881 .

ZHOU L H , FANG X M , HE R , et al . Deep reinforcement learning-based joint frame length and rate adaption for WLAN network [C ] // Proceedings of the 2023 IEEE 98th Vehicular Technology Conference (VTC2023-Fall) . Piscataway : IEEE Press , 2023 : 1 - 5 .

WANG D , HU M Q . Deep deterministic policy gradient with compatible critic network [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2023 , 34 ( 8 ): 4332 - 4344 .

AGGARWAL C C . Neural networks and deep learning: a textbook [M ] . Cham : Springer , 2018 .

HO J , JAIN A , ABBEEL P . Denoising diffusion probabilistic models [EB ] . 2020 .

SCHULMAN J , WOLSKI F , DHARIWAL P , et al . Proximal policy optimization algorithms [EB ] . 2017 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

车联网中基于迁移强化学习的跨域充电站推荐算法

基于D3QN的Wi-Fi网络智能调制方法

PD-TD3：高速公路场景下边路协同计算卸载策略

基于深度强化学习的多租户算网资源分配算法

不依赖位置坐标的室内Wi-Fi网络覆盖度量方法