On-Demand Asynchronous Localization for Underwater Sensor Networks

In this paper we consider the issue of localization in the context of underwater sensor networks which contain anchor nodes with perfect knowledge of their position, but asynchronous clocks. By taking advantage of a sequential transmission protocol and the broadcasting nature of the acoustic underwater medium, the entire network can be localized simultaneously with small overhead. Additionally, it can be initiated by any node at any time. Through extensive simulation and derivation of the Cramer-Rao Lower Bound (CRLB), we first verify the utility and performance of the algorithm, demonstrating that both initiator and passive nodes can achieve low-error positioning. We then implement the algorithm on an existing modem, and through tests performed within a pool and lake we have determined the accuracy and effectiveness of the algorithm in a true underwater environment.