Real-Time Relative Positioning with WSN

This paper addresses a scenario where a wireless sensor network (WSN) is used to implement a real-time relative positioning system. Its goal is to obtain relative coordinates of objects (the WSN nodes) in which the nodes are generally moving relatively to each other. To accomplish this, the distance among nodes is measured and its relative position is computed allowing the representation of their spatial distribution. The test platform uses Crossbow´s motes to which special ultrasound (US) boards have been adapted. The US is thus the process used to measure distances in this system. Some of the issues addressed by this work are: precise time-synchronization between nodes, distance measurement procedure, positioning algorithms and its implementation on the TinyOS, operating system used in the Crossbow´s motes. Test shave shown that object positioning could be obtained with errors below 2% and using a computing time of about 1s per network node, on average.