Localization of large scale underwater sensor networks based on recursive position estimation

Large scale underwater wireless sensor networks (UWSNs) have great applications in ocean exploration. The basic localization scheme is the recursive position estimation with hierarchical steps. To achieve accurate localization for each node and a rather high localization coverage, the localized nodes to be chosen as reference nodes are one of the critical steps. This paper proposed a new method to define the confidence value of a reference node based on covariance law. With the measured distance between nodes, the estimated position is more accurate throughout the network and can give a comprehensive localization error. Localization coverage and localization error are used to evaluate the performance. Compared with the previous methods, the simulation results show that the proposed method have great advantage in controlling error accumulation, it is suitable for high precision large scale UWSNs.