In this paper, we investigate the impact of environmental changes on Medium Access Control (MAC) and routing protocols for underwater acoustic networks. We carry out the evaluation using the ns2-Miracle network simulator and the WOSS extensions, which interface the simulator to the Bellhop ray tracing software. We further extend the simulator to take into account the change of environmental parameters during the day, and to be able to generate random realizations of surface waves. We start by discussing how the acoustic propagation pattern changes due to changing temperature conditions, and show the impact of such variability on the performance of two random access protocols, namely CSMA-ALOHA and DACAP. As CSMA-ALOHA proves best in our simulation, we consider this protocol in a converge-casting scenario, where all nodes have to deliver their data to a centrally placed sink. In this scenario, we show that keeping the routes fixed is not the best strategy because of time-varying propagation effects, and that even infrequent route updates (once every 3 hours) achieve much better results in terms of throughput, delivery delay, and average route length than static routes.
On the impact of the environment on MAC and routing in shallow water scenarios
CASARI, PAOLO;ZORZI, MICHELE
2011
Abstract
In this paper, we investigate the impact of environmental changes on Medium Access Control (MAC) and routing protocols for underwater acoustic networks. We carry out the evaluation using the ns2-Miracle network simulator and the WOSS extensions, which interface the simulator to the Bellhop ray tracing software. We further extend the simulator to take into account the change of environmental parameters during the day, and to be able to generate random realizations of surface waves. We start by discussing how the acoustic propagation pattern changes due to changing temperature conditions, and show the impact of such variability on the performance of two random access protocols, namely CSMA-ALOHA and DACAP. As CSMA-ALOHA proves best in our simulation, we consider this protocol in a converge-casting scenario, where all nodes have to deliver their data to a centrally placed sink. In this scenario, we show that keeping the routes fixed is not the best strategy because of time-varying propagation effects, and that even infrequent route updates (once every 3 hours) achieve much better results in terms of throughput, delivery delay, and average route length than static routes.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.