A new MAC based on RTT prediction for underwater acoustic networks.

Most existing medium access control (MAC) protocols in underwater acoustic networks (UANs) ignore the delay variance that affects the estimation accuracy of round trip time (RTT). We predict the RTTs using Bayesian dynamic linear model. Using the predicted RTTs, we dynamically adjust the length of time slots in MAC. Experimental results show that the predicted values can adapt quickly to the delay variance in the acoustic channel.