Energy-Aware Node Selection for Localization

This work presents node selection algorithms in a resource-constrained environment where the algorithms maintain a desirable geolocation accuracy while extending the tracking lifetime of the system. The sensor manager simply selects an active set of nodes for a given snapshot by maximizing a utility function under the geolocation constraint. The utility function serves as a surrogate for the effective network lifetime. Three utility functions are considered: energy consumption (EC), remaining energy (RE), and current lifetime (CL). A general h-horizon node selection algorithm is formulated for a generic utility function. Then, the node selection algorithms corresponding to the three utilities are tested for the myopic (h = 1) and h = 2 cases. The results indicate that the RE and CL utilities lead to longer effective network lifetimes than the EC utility.