A time-frequency analysis method based on Fourier Decomposition Method(FDM) has been introduced to improve the accuracy of time-frequency analysis of blasting vibration signals.Firstly,the original simulated signals were decomposed based on FDM theory,and the correlation coefficient and energy ratio between the modal component and the original signal were calculated,so as to screen the dominant components.Then,the Hilbert transform was applied to the screened components to obtain the time-frequency spectrum.At the same time,the original simulated signals were decomposed based on EMD,EEMD and CEEMDAN methods,and the modal mixing effect of the decomposition results was evaluated by using the energy ratio theory.In addition,the corresponding time-frequency spectrums were obtained by using Hilbert transform,and the time-frequency resolutions of the FDM,EMD,EEMD and CEEMDAN methods were compared.Finally,the proposed method was applied to time-frequency analysis of measured blasting vibration signals.The results show that FDM can effectively solve the problem of modal aliasing,and the decomposition results have a better time-frequency resolution by the Hilbert transform.The proposed method is advantageous to the acquisition of local features and the time-frequency-energy relations of the blasting vibration signals,so as to improve the accuracy of time-frequency analysis of tunnel blasting vibration signals.