Localization based on best spatial correlation distance mobility prediction for underwater wireless sensor networks

In order to reduce the communication cost while keeping the localization coverage and localization accuracy high, we propose a new localization scheme for underwater wireless sensor networks, i.e., localization based on best spatial correlation distance mobility prediction (LBMP). Nodes predict their mobility pattern by utilizing the spatial correlation between the mobility of underwater nodes and get located. In order to keep the localization error small, nodes with great computation ability calculate the best spatial correlation distance for the neighbor nodes to predict their mobility pattern. LBMP defines the confidence of a node to evaluate the accuracy of mobility pattern and location prediction. By controlling the value of the confidence threshold, LBMP can guarantee the quality of mobility pattern and location prediction. Simulation experiments show that, comparing to the scheme without the selection of best spatial correlation distance, LBMP has better performance in keeping relatively high localization coverage and localization accuracy while reducing communication cost apparently.