Multiscale zero-crossing statistics of intrinsic mode functions for white Gaussian noise

Abstract

In this paper, the statistical characteristics of zero-crossing intervals and zero-crossing amplitudes (the time interval and the absolute value of extrema between two successive zero-crossings) of intrinsic mode functions of white Gaussian noise are studied. Intrinsic mode functions are extracted by empirical mode decomposition method. Numerous simulations are conducted and the probability distribution functions of zero-crossing intervals and amplitudes are obtained. Simulation results are included. These findings are important to determine the statistical significance of IMFs. These white noise-only case statistical characteristics can be used for signal detection and/or signal/noise separation and an efficient noise reduction can be achieved.