Reference Anchor Selection and Global Optimized Solution for DV-Hop Localization in Wireless Sensor Networks

Localization is a fundamental issue in wireless sensor networks (WSN). Among typical range-free localization algorithms, distance vector-hop (DV-Hop) is preferred as it can localize unknown nodes with less than three neighbor anchors. However, the accuracy of DV-Hop localization is low due to its multi-hop nature and defective position-calculation procedure. This paper first investigates a neglected issue during position calculation, i.e., reference anchor selection problem in the third step of DV-Hop. Then in order to solve this problem, two new algorithms are proposed. As for the first algorithm named as RAS DV-Hop, since every anchor can act as a reference anchor, an unknown node can obtain several candidate positions, and the best candidate that has the most similar distance to anchors with the unknown node is chosen as the final estimated position. The second algorithm named as GOS DV-Hop provides a global optimized solution for the third step of DV-Hop, achieves better accuracy than RAS DV-Hop, but has higher computation complexity. Simulation results indicate that both proposed algorithms achieves better precision, compared to typical DV-Hop based algorithms.