Title: An improved prediction based strategy for target tracking in wireless sensor networks

Authors: Hanen Ahmadi; Ridha Bouallegue; Federico Viani; Andrea Massa

Addresses: Innov'COM, Supcom, University of Carthage/University of Tunis El Manar, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@Innov’COM), University of Carthage, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@UniTN), University of Trento, Trento, Italy ' Innov'COM, Supcom, University of Carthage/University of Tunis El Manar, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@Innov’COM), University of Carthage, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@UniTN), University of Trento, Trento, Italy ' Innov'COM, Supcom, University of Carthage/University of Tunis El Manar, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@Innov’COM), University of Carthage, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@UniTN), University of Trento, Trento, Italy ' Innov'COM, Supcom, University of Carthage/University of Tunis El Manar, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@Innov’COM), University of Carthage, Tunis, Tunisia; ELEDIA Research Center (ELEDIA@UniTN), University of Trento, Trento, Italy

Abstract: The indoor localisation of moving target in wireless sensor networks using received signal strength indicator (RSSI) is addressed in this paper. A novel location tracking algorithm which combines an ensemble learning method and Kalman filter is proposed. An ensemble-based regression tree using received signal strength method has been proposed to localise static sensor nodes. In this paper, this approach is employed to solve the complex relation between the RSSI behaviour and the target position. Then, the estimated location is introduced in the Kalman filter as the observed information, leading to more accurate state of the moving target. Experimental results show that the adopted solution achieves a high accuracy compared to localisation algorithms currently available in the literature.

Keywords: target tracking; localisation; wireless sensor network; WSN; machine learning; Kalman filter.

DOI: 10.1504/IJITST.2018.093667

International Journal of Internet Technology and Secured Transactions, 2018 Vol.8 No.3, pp.453 - 468

Received: 02 Mar 2017
Accepted: 29 May 2017
Published online: 31 Jul 2018 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article