Research on restraining the end effect of EMD based on grey prediction model

Empirical Mode Decomposition(EMD) is an advanced method for analyzing non-stationary signal, but there is an involved end issue in the course of getting two envelops of the data using spline interpolation. In this paper a novel method based on grey prediction model is proposed to restrain the end effect of empirical mode decomposition. In the grey prediction endpoint extension process, based on the theory of GM(1,1) prediction model, the endpoints are extended respectively to approach the variation tendency of the original data sequences. Simulation result shows that the proposed method can restrain end effect effectively and the parameters of grey prediction are easy to determine.