Localization of the offshore pollutant source in lakes using spatial-temporal filtering

This paper proposes a method to localize an offshore pollutant source in lakes. The concentration of pollutant is measured periodically over time by sensor nodes distributed spatially. The location of the pollutant source is estimated by using unscented Kalman (UKF) spatial-temporal filtering approach when new temporal and spatial concentration measurements are obtained. In simulations, we compare the performance of the spatial-temporal filtering method with that of the only temporal filtering method. In addition, we analyze the localization effects caused by noises and initial values. Simulation results show that the spatial-temporal filtering method has smaller localization error and better robustness. When the initial value is close to the actual value, both methods perform accurate localization. While the initial value is far from the actual value, the spatial-temporal filtering method gets more accurate localization result.