关键词: 经验模态分解, 模态混叠, 端点效应

Abstract: Aiming at avoiding the existing problem of end effects and mode mixing in empirical mode decomposition, a new method is proposed. By using the method of mirror image extending to extend the signal data at both ends, then processing by cosine window function, the influence of end effects on decomposition results is solved. Using the method of high-frequency harmonics, it combines with the masking signal to inhibit the mode mixing of empirical mode decomposition. Simulation results indicate that the new method is effective and feasible.

Key words: empirical mode decomposition, mode mixing, end effects