Research on the method of features extraction for non-stationary signal based on local mean decomposition

For solving the problem of features extraction on non-stationary transient impact signal, a new method of features extraction, namely, the local mean decomposition (LMD), is introduced in this paper. The new method decomposes such signals into a set of functions, each of which is the product of an envelope signal and a frequency modulated signal from which a time-varying instantaneous frequency can be derived. The LMD method can be used to analyse non-stationary transient impact signal. Compared with wavelet decomposition, this algorithm not only avoids the frequency overlapping in the wavelet decomposition, but also overcomes the difficulty in selecting elementary wavelet. The numerical simulation shows the feasibility of the method. The example analysis of transient signal shows the reliability of the method.