ACOUFIND: Acoustic ad-hoc network system for trapped person detection

Most existing systems for finding live survivors under rubble in case of a disaster are based on RF or seismic detection. These systems generally do not provide accurate location of the victim, have high set-up and scan time and are prone to noise and interference. In this paper we describe an acoustic detection system that provides an accurate location of live persons under rubble. The system is based on a smartphone application that records the cries for help of the survivors and sounds nearby. The application is installed on smartphones of rescue team members positioned around the site. Recorded sound is transferred to a central processor that applies an acoustic location algorithm and provides an estimation of survivors location. The prototype system was tested in the Israel Home Front Command training site. Location accuracy measured for only three acoustic records around the survivor was within two meters on a 50 meters diameter site.