Research On Seismic Source Location Technology Based On Multi-seismic Sensors

In the seismic source detection system, the seismic sensors are used to acquireseismic signal, then the location of moving target can be obtained through the analysis ofthe acquisition seismic signal. This passive detection technology based on seismicsensors has great application value in both military and civil aspect. However, thelocation accuracy is still a difficulty of unattended ground sensors technology. Therefore,it is essential to study on seismic source location technology which is based on seismicsensors.Firstly, this paper studies the seismic wave propagation theory and determines todetect Rayleigh wave. According to the characteristics of seismic wave propagation, theseismic signal acquisition system is designed. Then, field experiment is described in theterms of experimental instrument, experimental field and experimental contents.The location algorithm is based on the time difference of arrival(TDOA), Timedelay estimation is the key problem of TDOA. General Cross Correlation(GCC) withdifferent weighted functions and Higher-Order Cumulant method are comparedaccording to the analysis of different signal-to-noise ratio footstep signals. Then, the timedelay estimation method which is more suitable for footstep signal is selected and thelocation algorithm based on empirical mode decomposition(EMD) is presented. Theproposed method has proved effective through experimental data processing and resultsanalysis.Compared with the classical location method, the location algorithm based onempirical mode decomposition can improve the location accuracy. But, the locationaccuracy of this method will be affected and the cross-correlation peak is not obvious ina complex environment. In order to solve this problem, the seismic source locationmethod based on seismic signal characteristic frequency is proposed. The wavelet packetis adopted to make de-noising pretreatment, then, the local mean decomposition(LMD)is applied to analyze seismic signal. The components are analyzed in time domain andfrequency domain. The seismic source location can be achieved according to timedifference of characteristic frequency arrival. The proposed method has proved effectivethrough experimental data processing and results analysis.