Deploying Federated Learning in Large-Scale Cellular Networks: Spatial Convergence Analysis

计算机科学 上传 蜂窝网络 数据传输 人口 传输(电信) 无线 延迟(音频) 计算机网络 分布式计算 电信 操作系统 社会学 人口学
作者
Zhenyi Lin,Xiaoyang Li,Vincent K. N. Lau,Yi Gong,Kaibin Huang
出处
期刊:IEEE Transactions on Wireless Communications [Institute of Electrical and Electronics Engineers]
卷期号:21 (3): 1542-1556 被引量:30
标识
DOI:10.1109/twc.2021.3104834
摘要

The deployment of federated learning in a wireless network, called federated edge learning (FEEL), exploits low-latency access to distributed mobile data to efficiently train an AI model while preserving data privacy. In this work, we study the spatial (i.e., spatially averaged) learning performance of FEEL deployed in a large-scale cellular network with spatially random distributed devices. Both the schemes of digital and analog transmission are considered, providing support of error-free uploading and over-the-air aggregation of local model updates by devices. The derived spatial convergence rate for digital transmission is found to be constrained by a limited number of active devices regardless of device density and converges to the ground-true rate exponentially fast as the number grows. The population of active devices depends on network parameters such as processing gain and signal-to-interference threshold for decoding. On the other hand, the limit does not exist for uncoded analog transmission. In this case, the spatial convergence rate is slowed down due to the direct exposure of signals to the perturbation of inter-cell interference. Nevertheless, the effect diminishes when devices are dense as interference is averaged out by aggressive over-the-air aggregation. In terms of learning latency (in second), analog transmission is preferred to the digital scheme as the former dramatically reduces multi-access latency by enabling simultaneous access.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
更新
PDF的下载单位、IP信息已删除 (2025-6-4)

科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
刚刚
1秒前
1秒前
1秒前
1秒前
fduqyy发布了新的文献求助10
2秒前
浮沉完成签到,获得积分10
2秒前
2秒前
Miracle完成签到,获得积分10
4秒前
赖氨酸发布了新的文献求助10
4秒前
5秒前
无花果应助WBTT采纳,获得10
5秒前
明理丹云完成签到,获得积分20
5秒前
Lotus完成签到,获得积分10
6秒前
SciGPT应助笑点低的沛蓝采纳,获得10
6秒前
爆米花应助fduqyy采纳,获得10
6秒前
今后应助欢欢欢乐乐乐乐采纳,获得10
6秒前
在水一方应助勤恳的惋庭采纳,获得10
6秒前
single完成签到,获得积分10
6秒前
小青椒应助IU秋阳采纳,获得30
7秒前
打打应助Zhang采纳,获得10
7秒前
8秒前
大模型应助ssy采纳,获得10
8秒前
张婷发布了新的文献求助10
8秒前
无情的绮彤完成签到,获得积分10
8秒前
9秒前
碧蓝白玉完成签到,获得积分10
9秒前
9秒前
9秒前
完美世界应助Jeff采纳,获得30
9秒前
科研通AI2S应助姚俊采纳,获得10
9秒前
10秒前
11秒前
11秒前
12秒前
pass发布了新的文献求助10
12秒前
微毒麻醉完成签到,获得积分10
12秒前
wei发布了新的文献求助10
13秒前
糖醋花孙米完成签到,获得积分10
13秒前
不可思议发布了新的文献求助10
13秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
On the Angular Distribution in Nuclear Reactions and Coincidence Measurements 1000
Vertébrés continentaux du Crétacé supérieur de Provence (Sud-Est de la France) 600
A complete Carnosaur Skeleton From Zigong, Sichuan- Yangchuanosaurus Hepingensis 四川自贡一完整肉食龙化石-和平永川龙 600
FUNDAMENTAL STUDY OF ADAPTIVE CONTROL SYSTEMS 500
微纳米加工技术及其应用 500
Nanoelectronics and Information Technology: Advanced Electronic Materials and Novel Devices 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 工程类 有机化学 生物化学 物理 纳米技术 计算机科学 内科学 化学工程 复合材料 物理化学 基因 遗传学 催化作用 冶金 量子力学 光电子学
热门帖子
关注 科研通微信公众号,转发送积分 5308126
求助须知:如何正确求助?哪些是违规求助? 4453339
关于积分的说明 13857031
捐赠科研通 4341040
什么是DOI,文献DOI怎么找? 2383601
邀请新用户注册赠送积分活动 1378277
关于科研通互助平台的介绍 1346269