There a lot of time–frequency methods used for studying the time–frequency distribution of the nonlinear and non-stationary signal, such as short time Fourier transform (STFT) (Zhang and Bao, 1998), Wavelet transform (WT) (Hu et al., 2009), S-transform (ST) (Mohammad, 2009, Strockwell, 2007, Strockwell et al., 1996 and Wang and Orchard, 2009), Wigner–Ville distribution (WVD) (Chen, 2007 and Lokenath, 2002), smoothed pseudo Wigner distribution (SPWD) (Qiao, 2010), matching pursuit (MP) (Chen et al., 2007 and Stéphane and Zhang, 1993), adaptive optimum kernel time–frequency representation (AOK) (Douglas and Richard, 1995 and Liu et al., 2008), and so on. Each method mentioned above has its own advantages and disadvantages. Hilbert–Huang transform (HHT) proposed by Huang et al., 1996, Huang et al., -There are a lot of time–frequency methods used for studying the time–frequency distribution of the nonlinear and non-stationary signal, such as short time Fourier transform (STFT) (Zhang and Bao, 1998), Wavelet transform (WT) (Hu et al., 2009), S-transform (ST) (Mohammad, 2009, Strockwell, 2007, Strockwell et al., 1996 and Wang and Orchard, 2009), Wigner–Ville distribution (WVD) (Chen, 2007 and Lokenath, 2002), smoothed pseudo Wigner distribution (SPWD) (Qiao, 2010), matching pursuit (MP) (Chen et al., 2007 and Stéphane and Zhang, 1993), adaptive optimum kernel time–frequency representation (AOK) (Douglas and Richard, 1995 and Liu et al., 2008), and so on. Each method mentioned above has its own advantages and disadvantages. Hilbert–Huang transform (HHT) proposed by Huang et al., 1996, Huang et al.,