Localization in a Sparse Wireless Sensor Network Using Pedometer and Communication Ranging Measurements

Localization of sensor nodes is an essential requirement for many wireless sensor network applications. As equipping GPS receivers to every node are prohibitively expensive with hundred of nodes in the system, GPS-less localization approach is necessary. In this paper, we present a low cost system based on measurements from a pedometer and communication ranging between neighboring nodes. The information is fussed through a probability based localization algorithm. Unlike most of the existing methods that require good network connectivity, the proposed system works reasonably well in a sparse network. An experimental prototype has been developed and the practical results are used to evaluate the system's performance.