Mobile data traffic prediction based on empirical mode decomposition and kernel extreme learning machine

Abstract

In view of the nonlinear and non-stationary characteristics of mobile data traffic, this paper proposes a mobile data traffic prediction model based on empirical mode decomposition (EMD) and kernel extreme learning machine (KELM). It uses EMD algorithm to decompose the mobile data traffic to obtain a series of Intrinsic Mode Function (IMF) with distinct frequency component. KELM is proposed for modeling and prediction for each IMF, based on which the mobile data traffic can be predicted by aggregating the predicted results of all KELM. Real mobile data traffic is used to verify the feasibility and effectiveness of the proposed method in this paper.