基于动态扩散卷积交互图神经网络的网络流量预测

现有的网络流量预测模型存在着泛化能力弱和预测准确率低等问题，为了解决此问题，提出了一种结合动态扩散卷积模块和卷积交互模块的预测模型。动态扩散卷积模块可以提取网络流量中复杂的空间特征和动态特性，而卷积交互模块则能捕获到流量中的时间特征，两者的有机结合可以有效预测网络中的流量。通过与其他网络流量预测模型在美国能源科学网（ESnet）流量数据上进行对比实验，验证了提出的动态扩散卷积交互图神经网络模型（DDCIGNN）的有效性。实验结果表明，DDCIGNN模型的均方根误差（RMSE）在最好的情况下优化了大约13.0％，说明该模型能够进行更有效的网络流量预测。

Network traffic prediction based on dynamic diffusion convolutional interaction graph neural network

The existing network traffic prediction models have problems such as weak generalization ability and low prediction accuracy. To solve this problem, this paper proposed a prediction model combining dynamic diffusion convolution module and convolution interaction module. The dynamic diffusion convolution module could extract the complex spatial and dynamic characteristics of network traffic, while the convolution interaction module could capture the temporal characteristics of the traffic. The two organic combination could effectively predict the traffic in the network. This paper verified the effectiveness of the proposed dynamic diffusion convolutional interaction graph neural network(DDCIGNN) model by comparative experiments with other network traffic prediction models on the flow data of the US energy science network(ESnet). Experimental results show that the root mean square error(RMSE) of the DDCIGNN model is optimized by about 13.0% in the best case, which indicates that the model can perform better in network traffic prediction.