Adaptive Graph Convolutional Recurrent Neural Networks for System-Level Mobile Traffic Forecasting  

作　　者：Yi Zhang Min Zhang Yihan Gui Yu Wang Hong Zhu Wenbin Chen Danshi Wang 

机构地区：[1]Beijing University of Posts and Telecommunications,Beijing 100876,China [2]The Intelligent Network Innovation Center of Chinaunicom,Beijing 100048,China

出　　处：《China Communications》2023年第10期200-211,共12页中国通信（英文版）

基　　金：supported by the National Natural Science Foundation of China(61975020,62171053)。

摘　　要：Accurate traffic pattern prediction in largescale networks is of great importance for intelligent system management and automatic resource allocation.System-level mobile traffic forecasting has significant challenges due to the tremendous temporal and spatial dynamics introduced by diverse Internet user behaviors and frequent traffic migration.Spatialtemporal graph modeling is an efficient approach for analyzing the spatial relations and temporal trends of mobile traffic in a large system.Previous research may not reflect the optimal dependency by ignoring inter-base station dependency or pre-determining the explicit geological distance as the interrelationship of base stations.To overcome the limitations of graph structure,this study proposes an adaptive graph convolutional network(AGCN)that captures the latent spatial dependency by developing self-adaptive dependency matrices and acquires temporal dependency using recurrent neural networks.Evaluated on two mobile network datasets,the experimental results demonstrate that this method outperforms other baselines and reduces the mean absolute error by 3.7%and 5.6%compared to time-series based approaches.

关 键 词：adaptive graph convolutional network mobile traffic prediction spatial-temporal dependence 

分 类 号：TN929.5[电子电信—通信与信息系统] TP183[电子电信—信息与通信工程]