Empirical mode decomposition (EMD) has recently been pioneered by Huang et al. for adaptively representing non-stationary signals as sums of zero-mean amplitude modulation frequency modulation components. The traditional EMD algorithm adopts the cubic spline interpolation as an effective tool processing non-stationary signal, but it cannot effectively extract the characteristic frequencies from a highly non-stationary signal, and the overshoots and the undershoots may become a common phenomenon during the decomposition process. In order to solve the problem, we presents the piecewise linear fractal interpolation as the spline interpolating. Finally, we will use the simulation signal to verify the effectiveness of the improved EMD.