Evaluating alternative approaches to mobile object localization in wireless sensor networks with passive architecture

In this study, we evaluate the performance of three types of techniques, namely neural based, Kalman filter based and trilateration based techniques, having been proposed to tackle the problem of real-time mobile sensor node tracking in a wireless sensor network with passive architecture. To investigate the performance of the aforementioned techniques under real-world circumstances, a small-scale wireless sensor network is deployed in an environment prone to multiple noise sources, multi-path and signal attenuation phenomena. The network makes use of a 433 MHz MICA2 based Cricket platform, which is comprised of 6 Cricket motes, at least one of which is mobile. The network utilizes a passive architecture in which any mobile mote receives the Beacon signals to localize itself. Subsequently, a neural based approach is compared with a trilateration and a Kalman filter based technique. The results obtained corroborate the efficiency and advanced performance of the neural based approach.

► In this study we compare three alternative approaches to localize mobile wireless sensors in an architecturally passive wireless sensor network.
► To investigate the performance of the techniques under real-world circumstances, a small-scale wireless sensor network is deployed in a noisy environment.
► The network makes use of a 433 MHz MICA2 based Cricket platform, which is comprised of 6 Cricket motes, at least one of which is mobile.
► The results obtained corroborate the efficiency and advanced performance of the neural based approach.