The Research on Phase Spectrum Decomposition Mechanism based on One-Side EMD

Aiming at the problem of Hilbert transformation on mode decomposition for non-Bedrosian signal, this paper proposes a phase spectrum decomposition method based on one-side empirical mode decomposition (EMD) which realizes the straight analysis for non-Bedrosian signal through Hilbert transformation. This paper first proposes the main features of the signal which meets the requirement of EMD, introducing the principle of phase spectrum superposition and one-side EMD to expand EMD application into phase spectrum scope. Then the frequencies relationship between time-amplitude coordinate and phase spectrum are deduced. Finally the process procedures as Hilbert transformation, one-side EMD phase spectrum decomposition and frequency analysis are constructed. From simulation testing, it is proved that one-side EMD phase spectrum decomposition mechanism could conquer the problem of Hilbert transformation for non-Bedrosian signal processing and be higher on stability and accuracy.