Discrete probabilistic localization of wireless sensor networks

Localizing sensor nodes is critical in the context of wireless sensor networks applications. It has been shown that, for certain applications, low overhead discrete localization achieves comparable results to costly fine localization. This paper presents a discrete and probabilistic localization method that requires no transmission overhead from the sensor nodes. The method is based on Kalman estimation that iteratively predicts and updates the position of a node with respect to a mobile reference. Simulations show that the method converges to the true position in a relatively short time with an average location accuracy of 91.1%.