Improving empirical mode decomposition with an optimized piecewise cubic Hermite interpolation method

Empirical mode decomposition (EMD) is an adaptive method for analyzing non-stationary time series derived from linear and nonlinear systems. But the upper and lower envelopes fitted by cubic spline (CS) interpolation may often occur overshoots. In this paper, a novel envelope fitting method based on the optimized piecewise cubic Hermite (OPCH) interpolation is developed. Taking the difference between extreme as the cost function, chaos particle swarm optimization (CPSO) method is used to optimize the derivatives of the interpolation nodes. The flattest envelope with the optimized derivatives can overcome the overshoots well. Some numerical experiments conclude this paper, and comparisons are carried out with the classical EMD.