Federated Deep Reinforcement Learning-Based Intelligent Channel Access in Dense Wi-Fi Deployments

📅 2024-09-02
🏛️ arXiv.org
📈 Citations: 0
Influential: 0
📄 PDF
🤖 AI Summary
In dense Wi-Fi scenarios, the conventional CSMA/CA mechanism suffers from high packet collision rates and excessive MAC-layer latency. To address these challenges, this paper proposes an intelligent channel access method jointly driven by Federated Learning (FL) and Deep Deterministic Policy Gradient (DDPG). We introduce a novel FL-DDPG co-design architecture, incorporating a lightweight training pruning strategy and an adaptive weighted aggregation algorithm to preserve user privacy and reduce communication overhead while enhancing adaptability to dynamic channel conditions. Extensive NS-3-AI simulations demonstrate that, under static conditions, the proposed method reduces average MAC latency by 25.24%; under dynamic conditions, it achieves 25.72% and 45.9% improvements over A-FRL and DRL baselines, respectively. These results substantiate significant gains in multi-AP concurrent access efficiency and system robustness.

Technology Category

Application Category

📝 Abstract
The IEEE 802.11 MAC layer utilizes the Carrier Sense Multiple Access with Collision Avoidance (CSMA/CA) mechanism for channel contention, but dense Wi-Fi deployments often cause high collision rates. To address this, this paper proposes an intelligent channel contention access mechanism that combines Federated Learning (FL) and Deep Deterministic Policy Gradient (DDPG) algorithms. We introduce a training pruning strategy and a weight aggregation algorithm to enhance model efficiency and reduce MAC delay. Using the NS3-AI framework, simulations show our method reduces average MAC delay by 25.24% in static scenarios and outperforms A-FRL and DRL by 25.72% and 45.9% in dynamic environments, respectively.
Problem

Research questions and friction points this paper is trying to address.

Wi-Fi Dense Areas
Signal Collision
Network Efficiency Degradation
Innovation

Methods, ideas, or system contributions that make the work stand out.

Federated Learning
Deep Deterministic Policy Gradient
Channel Optimization
🔎 Similar Papers
No similar papers found.
X
Xinyang Du
Key Lab of Information Coding & Transmission, Southwest Jiaotong University, Chengdu 610031, China
X
Xuming Fang
Key Lab of Information Coding & Transmission, Southwest Jiaotong University, Chengdu 610031, China
R
Rong He
Key Lab of Information Coding & Transmission, Southwest Jiaotong University, Chengdu 610031, China
L
Li Yan
Key Lab of Information Coding & Transmission, Southwest Jiaotong University, Chengdu 610031, China
L
Liuming Lu
Guangdong OPPO Mobile Telecommunications Corp., Ltd, GuangDong 523859, China
C
Chaoming Luo
Guangdong OPPO Mobile Telecommunications Corp., Ltd, GuangDong 523859, China