On the frequency resolution of improved empirical mode decomposition method

Empirical Mode Decomposition, which constitutes the core of Hilbert Huang Transform, is an efficient method in analysis of non-stationary and non-linear signals due to its data-driven nature. Nevertheless, the original method imposes a limitation on the frequency resolution and also suffers due to end effects. In this paper, through analyses of synthetic and actual vibration data, it is shown that better frequency resolution beyond the limitation imposed by the original method can be achieved by modifying the local mean approximation and boundary values of the original algorithm.