Distributed direction-based localization in wireless sensor networks

Localization is an important issue in wireless sensor networks (WSNs). Several applications in WSNs can tolerate the diverse levels of accuracy in geographic information such as physical location. Besides, direction information is sufficient for numerous applications in WSNs. Thus, the paper develops a direction-based localization scheme, DLS, in which a node can be aware of its relative direction to a sink rather than its physical absolute location without requiring additional equipments. DLS not only takes the spatial locality property into account, but also exploits the anchor deployment strategy to benefit sensor direction learning. Additionally, DLS can more precisely position sensors around the boundaries with the aid of the dual-direction coordinate system. Experimental results validate the practicability of DLS and show that DLS is able to achieve well accurate direction estimates.