Maximum Likelihood Localization of a Diffusive Point Source Using Binary Observations

In this paper, we investigate the problem of localization of a diffusive point source of gas based on binary observations provided by a distributed chemical sensor network. We motivate the use of the maximum likelihood (ML) estimator for this scenario by proving that it is consistent and asymptotically efficient, when the density of the sensors becomes infinite. We utilize two different estimation approaches, ML estimation based on all the observations (i.e., batch processing) and approximate ML estimation using only new observations and the previous estimate (i.e., real time processing). The performance of these estimators is compared with theoretical bounds and is shown to achieve excellent performance, even with a finite number of sensors