The Application Of Fractal Theory In P-wave Detection

The accurate P-wave detection is of fundamental importance in early warning. The traditional P-wave detecting methods are based on variations of the amplitude or frequency. The accuracy of traditional methods largely depends on the signal to noise ratio (SNR) of seismic events. Larger deviations of P-wave detection results and false triggering would happen under low SNR events. The application of fractal theory into the seismic signal triggering and P-wave detection is effective. In the fractal-based method,the geometrical characteristics of seismic signal and noise signal is described by the fractal dimension, and the seismic signal can be distinguished from the noise signal by the fractal dimension distance and the variation of fractal dimension of seismic signal depicts the P-wave arrival. In this paper, a STAFD/LTAFD method was proposed. The impact of parameters on P-wave detection results was analyzed. Using K-NET seismic data. STAFD/LTAFD method and traditional methods are used to detect the P-wave arrival under different SNR levels. Results show that. the STAFD/LTAFD method outperforms traditional methods with higher robustness and lower time deviation. At the same time, based on the data collected from earthquake early warning system, fractal characteristic modes of different noises were established and the fractal dimension distance between established modes and signals to be determined was calculated. Results show that. lower fractal dimension distance demonstrated the similarity between signals to be determined and characteristic modes. The above results demonstrate that it is feasible to apply the fractal theory into the research of seismic signal triggering and its P-wave detection.