Towards an automatic seismic localization of human footsteps
Today, many companies or government institutions are required to monitor outdoor areas of office buildings, frontiers or restricted areas. In this paper we present an automated algorithm to monitor and to localize human footsteps with the help of their excited seismic waves. The preliminary goal of the data processing is the calculation of traveltime differences for each receiver pair of an arbitrary array of geophones. The data are filtered with a band-pass filter, which improves the signalto-noise ratio significantly. A short term average / long term average (sta / lta) picker determines the time of the first breaks. After cutting out time windows around the picked arrival times, the traveltime differences are than calculated for each receiver pair using cross-correlation. Finally, a grid search algorithm is used to invert the traveltime differences into locations of the footstep. Therefore, the propagation velocity of the Rayleigh wave has to be approximated. In one field measurement reliable and precise localizations were achieved. Footsteps on a fixed position were localized within an area of 30 m x 30 m with less than one metre deviation and paths on a straight line were reconstructed successfully.