Hybrid Approach for Localization in Anisotropic Sensor Networks

In many real-world applications including agricultural, meteorological, military applications, etc, localization techniques are widely used to estimate the geographic locations of sensor nodes based on the precision positions of a few anchors equipped with special hardware. Existing localization algorithms mainly try to improve their accuracy in position estimation by using various heuristic-based or mathematical techniques. Every node in the network follows the same technique to find its physical location. However, each individual method with its own strength can only outperform the others in some but not all nodes. Based on this observation, we develop a hybrid approach for the localization problem. Each node collects the same kind of information. By analysing the information, a node can decide what is the best localization algorithm to use. Different nodes can make their own decisions. Our simulation results reveal that the hybrid approach is effective that it outperforms existing algorithms. To the best of our knowledge, our work presents the first effort in solving the absolute localization problem by adopting a hybrid approach