Sniper localization using passive acoustic measurements over an ad-hoc sensor array

Sniper localization has become a very critical problem in modern day land warfare. A gunshot leaves many physical signatures which can be utilized to trace the direction and origin of it. Audio signals are by far the most reliable and affordable such signatures and have been utilized for the said task. In this paper, the characteristic shape of shock-wave waveforms along with Time of Arrival (ToA) information about shock-waves and muzzle blasts are used to find the trajectory and origin of the bullet. A three-dimensional geometry has been considered and the system has been designed to perform under constraints such as asynchronous sensors, weak muzzle detection and sensor self-location errors. The system presented here is applied on a set of data generated using a supersonic projectile. The current approach is shown to be an improvement over existing shock-wave based methods. The robustness of the system to measurement errors has also been analyzed.