Studies show that defected rolling element bearing brings out periodic impact force. Therefore, extracting the cyclic characteristics of the impact force is of significant importance for the bearing diagnosis. In the present study, a novel fault characteristic frequency extraction method is proposed based on the empirical mode decomposition (EMD) and the logarithm of the shorttime energy. Firstly, the intrinsic mode function (IMF) containing the fault is sifted by the empirical mode decomposition and correlation coefficient. Then the shorttime energy of the IMF is integrated and the natural logarithm is considered to obtain the energy curve. Finally, the fault characteristic frequency of the bearing is found by the spectral analysis of the energy curve. The effectiveness of the proposed method is verified by simulation and experiments. Comparison of the proposed method with Hilbert envelope method and energy operator method shows that the proposed method can highlight the fault characteristic frequency.