Xinyang Du, Xuming Fang, Rong He, Li Yan, Liuming Lu, Chaoming Luo
{"title":"密集 Wi-Fi 部署中基于联合深度强化学习的智能信道接入","authors":"Xinyang Du, Xuming Fang, Rong He, Li Yan, Liuming Lu, Chaoming Luo","doi":"arxiv-2409.01004","DOIUrl":null,"url":null,"abstract":"The IEEE 802.11 MAC layer utilizes the Carrier Sense Multiple Access with\nCollision Avoidance (CSMA/CA) mechanism for channel contention and access.\nHowever, in densely deployed Wi-Fi scenarios, intense competition may lead to\npacket collisions among users. Although many studies have used machine learning\nmethods to optimize channel contention and access mechanisms, most of them are\nbased on AP-centric single-agent models or distributed models, which still\nsuffer poor generalization and insensitivity to dynamic environments. To\naddress these challenges, this paper proposes an intelligent channel contention\naccess mechanism that combines Federated Learning (FL) and Deep Deterministic\nPolicy Gradient (DDPG) algorithms. Additionally, an FL model training pruning\nstrategy and weight aggregation algorithm are designed to enhance the\neffectiveness of training samples and reduce the average MAC delay. We evaluate\nand validate the proposed solution using NS3-AI framework. Simulation results\nshow that in static scenarios, our proposed scheme reduces the average MAC\ndelay by 25.24% compared to traditional FL algorithms. In dynamic scenarios, it\noutperforms Average Federated Reinforcement Learning (A-FRL) and distributed\nDeep Reinforcement Learning (DRL) algorithms by 25.72% and 45.9%, respectively.","PeriodicalId":501280,"journal":{"name":"arXiv - CS - Networking and Internet Architecture","volume":"5 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2024-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Federated Deep Reinforcement Learning-Based Intelligent Channel Access in Dense Wi-Fi Deployments\",\"authors\":\"Xinyang Du, Xuming Fang, Rong He, Li Yan, Liuming Lu, Chaoming Luo\",\"doi\":\"arxiv-2409.01004\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"The IEEE 802.11 MAC layer utilizes the Carrier Sense Multiple Access with\\nCollision Avoidance (CSMA/CA) mechanism for channel contention and access.\\nHowever, in densely deployed Wi-Fi scenarios, intense competition may lead to\\npacket collisions among users. Although many studies have used machine learning\\nmethods to optimize channel contention and access mechanisms, most of them are\\nbased on AP-centric single-agent models or distributed models, which still\\nsuffer poor generalization and insensitivity to dynamic environments. To\\naddress these challenges, this paper proposes an intelligent channel contention\\naccess mechanism that combines Federated Learning (FL) and Deep Deterministic\\nPolicy Gradient (DDPG) algorithms. Additionally, an FL model training pruning\\nstrategy and weight aggregation algorithm are designed to enhance the\\neffectiveness of training samples and reduce the average MAC delay. We evaluate\\nand validate the proposed solution using NS3-AI framework. Simulation results\\nshow that in static scenarios, our proposed scheme reduces the average MAC\\ndelay by 25.24% compared to traditional FL algorithms. In dynamic scenarios, it\\noutperforms Average Federated Reinforcement Learning (A-FRL) and distributed\\nDeep Reinforcement Learning (DRL) algorithms by 25.72% and 45.9%, respectively.\",\"PeriodicalId\":501280,\"journal\":{\"name\":\"arXiv - CS - Networking and Internet Architecture\",\"volume\":\"5 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-02\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - CS - Networking and Internet Architecture\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.01004\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - CS - Networking and Internet Architecture","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.01004","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
摘要
IEEE 802.11 MAC 层利用载波侦测多路访问与碰撞避免(CSMA/CA)机制进行信道争用和访问。然而,在密集部署的 Wi-Fi 场景中,激烈的竞争可能会导致用户之间的分组碰撞。虽然许多研究都使用机器学习方法来优化信道争用和接入机制,但大多数研究都是基于以接入点为中心的单个代理模型或分布式模型,这些模型仍然存在泛化能力差和对动态环境不敏感的问题。为了应对这些挑战,本文提出了一种智能信道争用和访问机制,该机制结合了联合学习(FL)和深度确定性策略梯度(DDPG)算法。此外,还设计了一种 FL 模型训练剪枝策略和权重聚合算法,以提高训练样本的有效性并降低平均 MAC 时延。我们使用 NS3-AI 框架对所提出的解决方案进行了评估和验证。仿真结果表明,在静态场景下,与传统的 FL 算法相比,我们提出的方案降低了 25.24% 的平均 MAC 延迟。在动态场景中,它优于平均联合强化学习(A-FRL)和分布式深度强化学习(DRL)算法,分别提高了 25.72% 和 45.9%。
Federated Deep Reinforcement Learning-Based Intelligent Channel Access in Dense Wi-Fi Deployments
The IEEE 802.11 MAC layer utilizes the Carrier Sense Multiple Access with
Collision Avoidance (CSMA/CA) mechanism for channel contention and access.
However, in densely deployed Wi-Fi scenarios, intense competition may lead to
packet collisions among users. Although many studies have used machine learning
methods to optimize channel contention and access mechanisms, most of them are
based on AP-centric single-agent models or distributed models, which still
suffer poor generalization and insensitivity to dynamic environments. To
address these challenges, this paper proposes an intelligent channel contention
access mechanism that combines Federated Learning (FL) and Deep Deterministic
Policy Gradient (DDPG) algorithms. Additionally, an FL model training pruning
strategy and weight aggregation algorithm are designed to enhance the
effectiveness of training samples and reduce the average MAC delay. We evaluate
and validate the proposed solution using NS3-AI framework. Simulation results
show that in static scenarios, our proposed scheme reduces the average MAC
delay by 25.24% compared to traditional FL algorithms. In dynamic scenarios, it
outperforms Average Federated Reinforcement Learning (A-FRL) and distributed
Deep Reinforcement Learning (DRL) algorithms by 25.72% and 45.9%, respectively.