A Low Complexity Algorithm for Sensor Localization

Abstract:
Locationing in sensor networks is an area that has received much attention over the last few years. In most applications of sensor networks, position-awareness at individual sensors is crucial for a successful operation of the sensor network. Estimation of relative node coordinates, based on measured distances between nodes in the network, has been suggested as a means to provide position-awareness in sensor networks. Localization algorithms, based on distance measurements, may be realized using complex optimization algorithms, estimating not only node coordinates, but also individual node clock-offsets. As sensor sizes decrease and power becomes a scarce resource, such complex locationing algorithms may not be feasible. In this paper, we present a distributed locationing algorithm, called the kernel algorithm. The kernel algorithm handles unknown clock-offsets through a time-difference-of- arrival approach and does not assume any synchronization among the nodes in the network. The algorithm offers robustness and accuracy while keeping complexity low by means of a divide and conquer approach. Numerical simulations verifies the performance and robustness of the proposed algorithm.