Fast Implementation of Orthogonal Empirical Mode Decomposition and Its Application into Singular Signal Detection

摘要

Since the Empirical Mode Decomposition (EMD) lacks strict orthogonality, the method of Orthogonal Empirical Mode Decomposition is innovationally proposed by this paper. The primary thought of this method is to obtain the Intrinsic Mode Function (IMF) and the residual function by auto-adaptive band-pass filtering. OEMD is proved to preserve strict orthogonality and completeness theoretically in this paper, and an algorithm to implement OEMD fast, IMF binary searching algorithm is built based on the point that the analytical band-pass filtering preserves perfect band-pass feature in the frequency domain. An application into the singular signal detection shows that OEMD successfully conquers mode aliasing, avoids the occurrence of fault mode, and is featured by fast computing speed. Furthermore, it has good performance to analyze non-stationary signal.