Localization in Ultra Wideband sensor networks using Tabu search

Localization of sensor nodes has become an important research topic in the area of Wireless Sensor Networks (WSN). Ever since the Federal Communications Commission (FCC) allowed the use of unlicensed Ultra-Wideband (UWB) communications, the accurate positioning and low cost characteristics of UWB made it a suitable physical layer standard for WSN. The localization of the sensor nodes can be done either by using direct method that gives exact solutions by solving a set of simultaneous non-linear equations or by using any non-linear optimization techniques. This paper proposes a novel and computationally efficient range-based localization method for WSN that uses Time of arrival (TOA) based ranging and Tabu Search (TS) based global optimization for the position estimation of sensor nodes. The sum of squared range errors of all neighboring anchor nodes is taken as the objective function for this optimization problem. Tabu search ensures that this objective function is minimized without being trapped into local optima. Computational experiments on data drawn from simulated WSNs show better convergence characteristics than the existing Simulated Annealing based WSN localization. Experiments are carried out for IEEE 802.15.4a channel model.