Event localization using sensor networks for sources with non-circular footprint

This paper investigates the use of wireless sensor networks for estimating the location of an event that emits a signal that propagates over a large region. The contribution of the paper is that it deals with events with non-circular footprint. In the overwhelming majority of the papers that deal with event detection and localization, it is assumed that the source has a circular footprint in the sense that the event generates a signal that propagates uniformly in all directions thus it can be detected by any sensor that is located in a circular disc around the source. In this paper, we consider a covariance detection algorithm and a non-uniform propagation model where the event footprint is not the usual circular disc. Another contribution of this paper is that it uses measurement correlation information for localizing the event. Finally, we present some preliminary results for localizing a plume source.