Abstract
The accurate distance estimation between anchor nodes and sensor nodes based on the hop count information is an important research problem in Wireless Sensor Networks (WSNs). The accuracy of the localization algorithm greatly impacts the performance of the location dependent protocols and applications such as routing and storage. In anisotropic network where the transmission path between anchor nodes and sensor nodes is not straight but curved, Euclidean distance between a pair of sensor nodes is not always proportional to their hop count distance, which undermines the assumption of many existing range free localization algorithms. Most of the existing works focus on increasing the accuracy by using the heuristics or mathematical techniques. In this paper, we proposed Friendly Anchor Selection Strategy (FASS). in which we show that the selection of friendly anchor nodes instead of using all the anchor nodes for accurate localization is a very important factor especially in anisotropic network. We first demonstrate that using all the anchor nodes in anisotropic network does not give accurate position estimation. Then we devised the method of selecting friendly anchor nodes for each sensor node in order to get accurate localization. Extensive simulation results show that, FASS can effectively select the friendly anchors from all the anchors in anisotropic network and therefore improve the localization accuracy by finally adjusting the average hop distance by utilizing the information from friendly anchors only.
