The Empirical Mode Decomposition Process of Non-stationary Signals

The Hilbert-Huang transform is a new method for analysing nonlinear and non-stationary data, and the empirical mode decomposition is key part of the method. The transform method raised by Norden E. Huang and others. The transform method is applied in many areas of signal analysis. In this paper, the precipitation data of Beijing is used as the study´data. The data is decomposed by empirical mode decomposition method. Then with Space-time index method, the author probes dynamical non-stationary in the original data and the decomposition data, and made research to empirical mode decomposition process of the non-stationary signals. Finally the conclusion is that the precipitation time series is truly containing the non-stable factor, and with the decomposition, non-stationarity is weaker and weaker in the experience mode decomposition, the low frequency component´s non-stationary is very weak.