关键词: 网络流量, 经验模态分解, 复杂网络, 果蝇优化算法

Abstract: Aiming at the shortcomings of the traditional network traffic prediction method in complex network environment, such as large prediction error and low precision, this paper proposes a network traffic prediction model based on EMFOA_LSSVM. EMD is used to extract the trend and detail features of the network traffic data, and the input and output matrix of the prediction model is constructed. The network traffic prediction in complex network environment is realized by MFOA_LSSVM. The experimental results show that, compared with MFOA_LSSVM, FOA_LSSVM, PSO_LSSVM and LSSVM, EMFOA_LSSVM has higher prediction accuracy and convergence speed, and provides the basis for network traffic prediction and early warning.

Key words: network traffic, empirical mode decomposition (EMD), complex network, fruit fly optimization algorithm