基于图卷积神经网络的分解可解释性的蜂窝网络流量预测模型

张子天; 温之馨; 诸葛斌; 吕智豪; 董黎刚; 蒋献

doi:10.11959/j.issn.1000-0801.2025120

您当前的位置：

首页 >

文章列表页 >

基于图卷积神经网络的分解可解释性的蜂窝网络流量预测模型

研究与开发 | 更新时间：2025-11-16

- 基于图卷积神经网络的分解可解释性的蜂窝网络流量预测模型
- A decomposable and interpretable cellular network traffic prediction model based on graph convolutional neural network
- 电信科学 2025年41卷第9期页码：93-107
- 作者机构：
  
  浙江工商大学，浙江杭州 310018
- 作者简介：
  
  [ "张子天（1988- ），男，博士，浙江工商大学信息与电子工程学院副研究员，主要研究方向为基于机器学习的网络流量预测与资源管理。" ]
  [ " 温之馨（1998- ），男，浙江工商大学信息与电子工程学院硕士生，主要研究方向为时间序列网络流量预测。" ]
  [ "诸葛斌（1976- ），男，博士，浙江工商大学信息与电子工程学院教授，主要研究方向为网络和通信技术、互联网技术和网络安全。" ]
  [ "吕智豪（1999- ），男，浙江工商大学信息与电子工程学院硕士生，主要研究方向为通信与网络。" ]
  [ "董黎刚（1972- ），男，博士，浙江工商大学信息与电子工程学院教授，主要研究方向为智能网络、在线教育。" ]
  [ "蒋献（1988- ），男，浙江工商大学信息与电子工程学院讲师、实验员，主要研究方向为在线教育。" ]
- 基金信息：
  
  国家自然科学基金资助项目(62301488;W2421086);浙江工商大学省属高校基本科研业务费项目(QRK23010)
- DOI：10.11959/j.issn.1000-0801.2025120
  中图分类号： TP301
- 收稿：2024-12-13，
  
  修回：2025-01-06，
  
  纸质出版：2025-09-20
- 稿件说明：
移动端阅览
张子天,温之馨,诸葛斌等.基于图卷积神经网络的分解可解释性的蜂窝网络流量预测模型[J].电信科学,2025,41(09):93-107.

ZHANG Zitian,WEN Zhixin,ZHUGE Bin,et al.A decomposable and interpretable cellular network traffic prediction model based on graph convolutional neural network[J].Telecommunications Science,2025,41(09):93-107.
张子天,温之馨,诸葛斌等.基于图卷积神经网络的分解可解释性的蜂窝网络流量预测模型[J].电信科学,2025,41(09):93-107. DOI： 10.11959/j.issn.1000-0801.2025120.

ZHANG Zitian,WEN Zhixin,ZHUGE Bin,et al.A decomposable and interpretable cellular network traffic prediction model based on graph convolutional neural network[J].Telecommunications Science,2025,41(09):93-107. DOI： 10.11959/j.issn.1000-0801.2025120.

摘要

随着智能互联应用在城市场景中的普及，城市网络流量的激增带来了新挑战。基站蜂窝网络中的流量预测是资源分配与调度等关键应用的核心，准确预测蜂窝流量对于高效分配网络资源尤为重要。然而在蜂窝网络流量预测研究中，复杂的城市蜂窝流量往往有着深层次的时间和空间特征需要挖掘，为了解决这个问题，提出一个基于图卷积神经网络的分解可解释性时空图卷积神经网络（DISTGCN）。该神经网络利用蜂窝网络流量分解以及时空相关性，提高了流量预测的准确性，同时分解后的流量特征增强了预测结果的可解释性。在真实的经典数据集上的实验结果表明，DISTGCN的预测性能优于传统深度学习预测模型和图神经网络模型。

Abstract

With the rapid proliferation of intelligent connected applications in urban environments

the dramatic increase in urban network traffic presents significant challenges. Accurate traffic prediction in cellular base station networks is critical for optimizing resource allocation and scheduling in various applications. However

the complexity of urban cellular traff

ic necessitates

the extraction of deep temporal and spatial features. To address this issue

a novel deep learning framework based on graph convolutional neural networks (GCN) was proposed

named decomposable and interpretable spatiotemporal graph convolutional neural network (DISTGCN). The decomposability of cellular network traffic and its spatiotemporal correlations were leveraged by DISTGCN to enhance prediction accuracy. Furthermore

the decomposed traffic features were shown to provide improved interpretability for the prediction results. Extensive experiments conducted on real-world benchmark datasets demonstrate that DISTGCN significantly outperforms conventional deep learning models and state-of-the-art GNN-based models in terms of prediction performance.

关键词

Keywords

references

Cisco . Cisco annual internet report (2018–2023) white paper [R ] . 2020 .

SRINIVASAN R . IEEE 802.16m evaluation methodology document (EMD) [J ] . IEEE 802.16m-08/ 004 r 3 , 2008 .

XU F L , LIN Y Y , HUANG J X , et al . Big data driven mobile traffic understanding and forecasting: a time series approach [J ] . IEEE Transactions on Services Computing , 2016 , 9 ( 5 ): 796 - 805 .

RIZWAN A , ARSHAD K , FIORANELLI F , et al . Mobile Internet activity estimation and analysis at high granularity: SVR model approach [C ] // Proceedings of the 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) . Piscataway : IEEE Press , 2018 : 1 - 7 .

程定国 , 曾浩洋 . 无线通信网络中流量分析技术综述 [J ] . 电讯技术 , 2023 , 63 ( 3 ): 441 - 447 .

CHENG D G , ZENG H Y . Overview of network traffic analysis techniques in wireless communication networks [J ] . Telecommunication Engineering , 2023 , 63 ( 3 ): 441 - 447 .

QIU C , ZHANG Y Y , FENG Z Y , et al . Spatio-temporal wireless traffic prediction with recurrent neural network [J ] . IEEE Wireless Communications Letters , 2018 , 7 ( 4 ): 554 - 557 .

KHAN A , FOUDA M M , DO D T , et al . Short-term traffic prediction using deep learning long short-term memory: taxonomy, applications, challenges, and future trends [J ] . IEEE Access , 2023 , 11 : 94371 - 94391 .

TRINH H D , GIUPPONI L , DINI P . Mobile traffic prediction from raw data using LSTM networks [C ] // Proceedings of the 2018 IEEE 29th Annual International Symposium on Personal, Indoor and Mobile Radio Communications (PIMRC) . Piscataway : IEEE Press , 2018 : 1827 - 1832 .

高勇 , 陆钱春 , 李锋 . 面向IP网络扩容应用的复杂网络流量预测方法 [J ] . 电信科学 , 2023 , 39 ( 9 ): 21 - 31 .

GAO Y , LU Q C , LI F . A complex network traffic prediction method for IP network expansion applications [J ] . Telecommunications Science , 2023 , 39 ( 9 ): 21 - 31 .

BHATTI U A , TANG H , WU G L , et al . Deep learning with graph convolutional networks: an overview and latest applications in computational intelligence [J ] . International Journal of Intelligent Systems , 2023 ( 1 ): 8342104 .

KOCHETKOVA I , KUSHCHAZLI A , BURTSEVA S , et al . Short-term mobile network traffic forecasting using seasonal ARIMA and holt-winters models [J ] . Future Internet , 2023 , 15 ( 9 ): 290 .

DABLAIN D , KRAWCZYK B , CHAWLA N V . DeepSMOTE: fusing deep learning and SMOTE for imbalanced data [J ] . IEEE Transactions on Neural Networks and Learning Systems , 2023 , 34 ( 9 ): 6390 - 6404 .

ROMAN I , SANTANA R , MENDIBURU A , et al . In-depth analysis of SVM kernel learning and its components [J ] . Neural Computing and Applications , 2021 , 33 ( 12 ): 6575 - 6594 .

HUANG C W , CHIANG C T , LI Q H . A study of deep learning networks on mobile traffic forecasting [C ] // Proceedings of the 2017 IEEE 28th Annual International Symposium on Personal, Indoor, and Mobile Radio Communications (PIMRC) . Piscataway : IEEE Press , 2017 : 1 - 6 .

BENGIO Y , SIMARD P , FRASCONI P . Learning long-term dependencies with gradient descent is difficult [J ] . IEEE Transactions on Neural Networks , 1994 , 5 ( 2 ): 157 - 166 .

DALGKITSIS A , LOUTA M , KARETSOS G T . Traffic forecasting in cellular networks using the LSTM RNN [C ] // Proceedings of the 22nd Pan-Hellenic Conference on Informatics . New York : ACM Press , 2018 : 28 - 33 .

SHIRI F M , PERUMAL T , MUSTAPHA N , et al . A comprehensive overview and comparative analysis on deep learning models: CNN, RNN, LSTM, GRU [EB ] . 2023 : 2305 .17473.

ZHANG C Y , PATRAS P . Long-term mobile traffic forecasting using deep spatio-temporal neural networks [C ] // Proceedings of the Eighteenth ACM International Symposium on Mobile Ad Hoc Networking and Computing . New York : ACM Press , 2018 : 231 - 240 .

皇甫晓瑛 , 钱惠敏 , 黄敏 . 结合注意力机制的深度神经网络综述 [J ] . 计算机与现代化 , 2023 ( 2 ): 40 - 49, 57 .

HUANGFU X Y , QIAN H M , HUANG M . A review of deep neural networks combined with attention mechanism [J ] . Computer and Modernization , 2023 ( 2 ): 40 - 49, 57 .

ZHANG Z W , YAN L , GU Y T . ST2T: a spatio-temporal transformer for cellular traffic prediction in digital twin systems [C ] // Proceedings of the 2023 IEEE 6th International Conference on Electronic Information and Communication Technology (ICEICT) . Piscataway : IEEE Press , 2023 : 1112 - 1117 .

WANG X , ZHOU Z M , XIAO F , et al . Spatio-temporal analysis and prediction of cellular traffic in metropolis [J ] . IEEE Transactions on Mobile Computing , 2019 , 18 ( 9 ): 2190 - 2202 .

ZHANG H Q , LU G Q , ZHAN M M , et al . Semi-supervised classification of graph convolutional networks with Laplacian rank constraints [J ] . Neural Processing Letters , 2022 , 54 ( 4 ): 2645 - 2656 .

YU B , YIN H T , ZHU Z X . Spatio-temporal graph convolutional networks: a deep learning framework for traffic forecasting [EB ] . 2017 .

GUO S N , LIN Y F , WAN H Y , et al . Learning dynamics and heterogeneity of spatial-temporal graph data for traffic forecasting [J ] . IEEE Transactions on Knowledge and Data Engineering , 2022 , 34 ( 11 ): 5415 - 5428 .

YAO Y , GU B , SU Z , et al . MVSTGN: a multi-view spatial-temporal graph network for cellular traffic prediction [J ] . IEEE Transactions on Mobile Computing , 2023 , 22 ( 5 ): 2837 - 2849 .

BARLACCHI G , DE NADAI M , LARCHER R , et al . A multi-source dataset of urban life in the city of Milan and the Province of Trentino [J ] . Scientific Data , 2015 , 2 : 150055 .

SACCENTI E , HENDRIKS M H W B , SMILDE A K . Corruption of the Pearson correlation coefficient by measurement error and its estimation, bias, and correction under different error models [J ] . Scientific Reports , 2020 , 10 : 438 .

PARK H , NOH J , HAM B . Learning memory-guided normality for anomaly detection [C ] // Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Piscataway : IEEE Press , 2020 : 14372 - 14381 .

CHUNG J , GULCEHRE C , CHO K , et al . Empirical evaluation of gated recurrent neural networks on sequence modeling [EB ] . 2014 .

ZHAO N , YE Z Y , PEI Y Y , et al . Spatial-temporal attention-convolution network for citywide cellular traffic prediction [J ] . IEEE Communications Letters , 2020 , 24 ( 11 ): 2532 - 2536 .

BAI L , YAO L N , LI C , et al . Adaptive graph convolutional recurrent network for traffic forecasting [J ] . Advances in Neural Information Processing Systems , 2020 , 33 : 17804 - 17815 .,

WANG X , YANG K X , WANG Z D , et al . Adaptive hybrid spatial-temporal graph neural network for cellular traffic prediction [C ] // Proceedings of the ICC 2023 - IEEE International Conference on Communications . Piscataway : IEEE Press , 2023 : 4026 - 4032 .

GAO L , GUAN L . Interpretability of machine learning: recent advances and future prospects [J ] . IEEE MultiMedia , 2023 , 30 ( 4 ): 105 - 118 .

WANG X , ZHAO J , ZHU L , et al . Adaptive multi-receptive field spatial-temporal graph convolutional network for traffic forecasting [C ] // Proceedings of the 2021 IEEE Global Communications Conference (GLOBECOM) . Piscataway : IEEE Press , 2021 : 1 - 7 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

水声通信物理层技术综述：从模型驱动到数据驱动

NOMA系统中基于深度学习的压缩感知脉冲噪声抑制方法

基于增强图神经网络和对比学习的复杂网络节点分类

基于深度学习和SVM-RFE的网络入侵检测模型

基于深度学习的多样化复杂网络影响力节点识别