Shooter localization by networked multichannel acoustic ground sensors

Gunshot acoustic localization for military and urban security systems has long been an important topic of research. In recent years the development of independent Unmanned Ground Sensors (UGS), interconnected through Wireless Sensor Networks (WSN), performing distributed cooperative localization, has grown in popularity. This paper proposes an asynchronous method of gunshot localization, performed by UGS, equipped with circular microphone arrays. Each UGS in the WSN estimates the Direction of Arrival (DOA) of acoustic events and the time delay between these events. Fusion nodes perform event identification, accounting for outliers (e.g. target hit noise), and shooter localization, based on gathered event information and WSN geometry. The approach is tested on real signals, acquired at a shooting range, and succeeds in localizing the shooter's position with a mean accuracy of 0.87 meters for 30 shots.