Predominant frequency detection of seismic signal based on Gabor–Wigner transform for earthquake early warning systems

A few methods of time–frequency distributions (TFD) have been employed to analyze and detect the occurrence of the predominant frequencies in the seismic signal. Comprehensive joint time–frequency analysis of seismic signal using the well-known short-time Fourier transform (STFT), Gabor transform (GT), Wigner–Ville distribution (WVD), and Choi–Williams distribution (CWD) have been carried out to overcome the problem encountered upon the analysis of seismic signal with fast Fourier transform (FFT), which fails to provide temporal information of the seismic signal. Furthermore, the seismic detection method is proposed using a new time–frequency method, which is known as Gabor–Wigner transform (GWT), to achieve better time–frequency resolution by removing the cross-terms. The performance of the time–frequency distributions including Gabor–Wigner transform on the seismic signal has been quantified by Renyi entropy. The present analysis proves the efficacy of the GWT on seismic signal detection. The detection of predominant frequency facilitates in determining the earthquake magnitude. The suitability of the proposed GWT method has been investigated for earthquake early warning systems.