Caching for Doubly Selective Fading Channels via Model-Agnostic Meta-Reinforcement Learning

IF 4 3区 计算机科学 Q1 COMPUTER SCIENCE, INFORMATION SYSTEMS IEEE Systems Journal Pub Date : 2024-08-22 DOI:10.1109/JSYST.2024.3442958
Weibao He;Fasheng Zhou;Dong Tang;Fang Fang;Wei Chen
{"title":"Caching for Doubly Selective Fading Channels via Model-Agnostic Meta-Reinforcement Learning","authors":"Weibao He;Fasheng Zhou;Dong Tang;Fang Fang;Wei Chen","doi":"10.1109/JSYST.2024.3442958","DOIUrl":null,"url":null,"abstract":"Edge caching is expected to alleviate the traffic consumption in next-generation communications. In this article, we consider the transmission delay in wideband communications deteriorated by rapid user movements, where the frequency-selective wideband fading channels become fast time-varying and hence \n<italic>doubly-selective</i>\n due to the user movements. To preferably allocate the caching resource in such circumstance, we introduce a coordinated caching network and accordingly formulate an allocation problem. However, the formulated problem is shown to be NP-hard. By considering the extremely high computational complexity to solve the NP-hard problem by traditional optimization algorithm, and considering only a few samples can be obtained for each training instance due to shortened coherence-time in the dynamical doubly selective fading channels, we propose a model-agnostic meta-reinforcement learning method to address the formulated problem. Particularly, the proposed method can efficiently recognize the unstable mobile channels and accordingly cache to reduce the overall transmission delay while only requires a few training samples. Numerical simulations are performed to verify the effectiveness of the proposed method and results show that the proposed one outperforms the commonly adopted existing method of deep-deterministic-policy-gradient learning in terms of average delay and cache hit rate.","PeriodicalId":55017,"journal":{"name":"IEEE Systems Journal","volume":"18 3","pages":"1776-1785"},"PeriodicalIF":4.0000,"publicationDate":"2024-08-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"IEEE Systems Journal","FirstCategoryId":"94","ListUrlMain":"https://ieeexplore.ieee.org/document/10643614/","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INFORMATION SYSTEMS","Score":null,"Total":0}
引用次数: 0

Abstract

Edge caching is expected to alleviate the traffic consumption in next-generation communications. In this article, we consider the transmission delay in wideband communications deteriorated by rapid user movements, where the frequency-selective wideband fading channels become fast time-varying and hence doubly-selective due to the user movements. To preferably allocate the caching resource in such circumstance, we introduce a coordinated caching network and accordingly formulate an allocation problem. However, the formulated problem is shown to be NP-hard. By considering the extremely high computational complexity to solve the NP-hard problem by traditional optimization algorithm, and considering only a few samples can be obtained for each training instance due to shortened coherence-time in the dynamical doubly selective fading channels, we propose a model-agnostic meta-reinforcement learning method to address the formulated problem. Particularly, the proposed method can efficiently recognize the unstable mobile channels and accordingly cache to reduce the overall transmission delay while only requires a few training samples. Numerical simulations are performed to verify the effectiveness of the proposed method and results show that the proposed one outperforms the commonly adopted existing method of deep-deterministic-policy-gradient learning in terms of average delay and cache hit rate.
查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
通过模型诊断元强化学习实现双选择性衰减信道缓存
边缘缓存有望缓解下一代通信的流量消耗。在本文中,我们考虑了宽带通信中因用户快速移动而恶化的传输延迟问题,在这种情况下,频率选择性宽带衰落信道会因用户移动而变得快速时变,从而产生双重选择性。在这种情况下,为了更好地分配缓存资源,我们引入了协调缓存网络,并相应地提出了一个分配问题。然而,所提出的问题被证明是 NP 难的。考虑到用传统优化算法解决 NP 难问题的计算复杂度极高,并且由于动态双选择性衰落信道的相干时间缩短,每个训练实例只能获得少量样本,我们提出了一种与模型无关的元强化学习方法来解决所提出的问题。特别是,所提出的方法能有效识别不稳定的移动信道,并相应地缓存以减少整体传输延迟,同时只需要少量的训练样本。为了验证所提方法的有效性,我们进行了数值模拟,结果表明,所提方法在平均延迟和缓存命中率方面优于目前普遍采用的深度-确定性-策略梯度学习方法。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
IEEE Systems Journal
IEEE Systems Journal 工程技术-电信学
CiteScore
9.80
自引率
6.80%
发文量
572
审稿时长
4.9 months
期刊介绍: This publication provides a systems-level, focused forum for application-oriented manuscripts that address complex systems and system-of-systems of national and global significance. It intends to encourage and facilitate cooperation and interaction among IEEE Societies with systems-level and systems engineering interest, and to attract non-IEEE contributors and readers from around the globe. Our IEEE Systems Council job is to address issues in new ways that are not solvable in the domains of the existing IEEE or other societies or global organizations. These problems do not fit within traditional hierarchical boundaries. For example, disaster response such as that triggered by Hurricane Katrina, tsunamis, or current volcanic eruptions is not solvable by pure engineering solutions. We need to think about changing and enlarging the paradigm to include systems issues.
期刊最新文献
Relationship between emotional state and masticatory system function in a group of healthy volunteers aged 18-21. Table of Contents Front Cover Editorial IEEE Systems Council Information
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1