Burst-Sensitive Traffic Forecast via Multi-Property Personalized Fusion in Federated Learning
文献类型：期刊论文
| 作者 | Xue, Jingjing6,7; Sun, Sheng6; Liu, Min6,7,8; Wang, Yuwei6; Meng, Xuying6; Wang, Jingyuan1,2; Zhang, JunBo3; Xu, Ke4,5 |
| 刊名 | IEEE TRANSACTIONS ON MOBILE COMPUTING
 |
| 出版日期 | 2025-07-01 |
| 卷号 | 24期号:7页码:5598-5614 |
| 关键词 | Predictive models Forecasting Data models Market research Fluctuations Computational modeling Training Telecommunication traffic Adaptation models Mathematical models Network traffic forecast federated learning (FL) heterogeneous data multi-scale representation personalized fusion |
| ISSN号 | 1536-1233 |
| DOI | 10.1109/TMC.2025.3538871 |
| 英文摘要 | For distributed network traffic prediction with data localization and privacy protection, Federated Learning (FL) enables collaborative training without raw data exchange across Base Stations (BSs). Nevertheless, traffic across BSs exhibit inherently heterogeneous trend burst and smooth fluctuation properties, but existing FL methods model single-scale series from only one view, which cannot simultaneously capture diverse trend and fluctuation properties, especially distinct burst distributions. In this paper, we propose Personalized Federated Forecasting with Multi-property Self-fusion (P2FMS), which can represent multi-scale traffic properties from different views. With precise multi-property representations, a fusion-level prediction decision is learned for each client in a personalized manner to promptly sense traffic bursts and improve forecasting performance in non-IID settings. Specifically, P2FMS decomposes the traffic series into distinct time scales, based on which, we effectively extract closeness, period, and trend properties from different views. The closeness and period are embedded through global-view representations with spatial correlations, while non-stationary trends are individually fitted from the client-side view. Furthermore, a personalized combiner is designed to accurately quantify the proportion of general fluctuation raws (i.e., closeness and period) and specific trend property in predictions, which enables multi-property self-fusion for each client to accommodate heterogeneous traffic patterns and enhance prediction accuracy. Besides, an alternant training mechanism is introduced to optimize property representation and fusion modules with the convergence guarantee. Extensive experiments on real-world datasets show that P2FMS outperforms status quo methods in both prediction performance and convergence time. |
| 资助项目 | National Key Research and Development Program of China[2021YFB2900102] ; National Natural Science Foundation of China[62472410] ; National Natural Science Foundation of China[62072436] ; National Natural Science Foundation of China[72222022] ; National Natural Science Foundation of China[72171013] ; National Natural Science Foundation of China[7224210] ; National Science Fund for Distinguished Young Scholars of China[62425201] ; National Natural Science Foundation of China ; National Science Fund for Distinguished Young Scholars of China |
| WOS研究方向 | Computer Science ; Telecommunications |
| 语种 | 英语 |
| WOS记录号 | WOS:001504135300041 |
| 出版者 | IEEE COMPUTER SOC |
| 源URL | [http://119.78.100.204/handle/2XEOYT63/42339]  |
| 专题 | 中国科学院计算技术研究所期刊论文_英文 |
| 通讯作者 | Liu, Min |
| 作者单位 | 1.Beihang Univ, Sch Comp Sci & Engn, Beijing 100191, Peoples R China 2.Beihang Univ, MIIT Key Lab Data Intelligence & Management, Beijing 100191, Peoples R China 3.JD Technol, JD Intelligent Cities Res, JD iCity, Beijing 101111, Peoples R China 4.Tsinghua Univ, Dept Comp Sci & Technol, Beijing 100190, Peoples R China 5.Zhongguancun Lab, Beijing 100081, Peoples R China 6.Chinese Acad Sci, Inst Comp Technol, Beijing 100045, Peoples R China 7.Univ Chinese Acad Sci, Beijing 101408, Peoples R China 8.Zhongguancun Lab, Beijing 100190, Peoples R China |
| 推荐引用方式 GB/T 7714 | Xue, Jingjing,Sun, Sheng,Liu, Min,et al. Burst-Sensitive Traffic Forecast via Multi-Property Personalized Fusion in Federated Learning[J]. IEEE TRANSACTIONS ON MOBILE COMPUTING,2025,24(7):5598-5614. |
| APA | Xue, Jingjing.,Sun, Sheng.,Liu, Min.,Wang, Yuwei.,Meng, Xuying.,...&Xu, Ke.(2025).Burst-Sensitive Traffic Forecast via Multi-Property Personalized Fusion in Federated Learning.IEEE TRANSACTIONS ON MOBILE COMPUTING,24(7),5598-5614. |
| MLA | Xue, Jingjing,et al."Burst-Sensitive Traffic Forecast via Multi-Property Personalized Fusion in Federated Learning".IEEE TRANSACTIONS ON MOBILE COMPUTING 24.7(2025):5598-5614. |
入库方式： OAI收割
来源：计算技术研究所
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。