Deployment-Friendly Link Adaptation in Wireless Local-Area Network Based on On-Line Reinforcement Learning

被引：1

作者：

Chen, Jie ^{[1
]}

Ma, Juntao ^{[1
]}

He, Yihao ^{[1
]}

Wu, Gang ^{[1
]}

机构：

[1] Univ Elect Sci & Technol China, Natl Key Lab Wireless Commun, Chengdu 611731, Peoples R China

来源：

IEEE COMMUNICATIONS LETTERS | 2023年 / 27卷 / 12期

关键词：

Signal to noise ratio; Wireless communication; Fading channels; Error analysis; Heuristic algorithms; Stochastic processes; Standards; Wireless networks; fading channel; outer loop link adaptation; reinforcement learning; Q-learning;

D O I：

10.1109/LCOMM.2023.3327964

中图分类号：

TN [电子技术、通信技术];

学科分类号：

0809 ;

摘要：

In this letter, based on outer loop link adaptation (OLLA), we propose an adaptive OLLA algorithm (AOLLA), which calculates the OLLA offset adapted to the current wireless channel environment in real-time by utilizing frequency statistics of packet error within the last observation window. AOLLA algorithm contains numerous parameters that require manual tuning, which raises the deployment difficulty of the algorithm. An online reinforcement learning algorithm is designed to tune the parameters automatically and allow the AOLLA algorithm to rapidly deploy in different environments. We worked out an experimental validation by deploying the proposed algorithm on a software-defined radio hardware platform in three typical scenarios for deploying a wireless local-area network using IEEE 802.11ax standard with its original HESU packet format. Experimental results show that our designed algorithm has a 59.6% performance improvement compared to the original OLLA.

引用

页码：3424 / 3428

页数：5