Adaptive Modulation and Feedback Strategy for an Underwater Acoustic Link

As underwater acoustic channels are highly variable, both temporally and spatially, it is impossible to design a single communication scheme that works well everywhere and at all times. Adaptive modulation and coding (AMC) techniques can help determine the best communication scheme to use for a particular channel, but require an accurate model to predict communication performance. We propose a bit error rate (BER) estimation model that fuses domain knowledge with experimental data, yielding BER estimates that are consistent with experimental observations. Predictions from such a model are used to drive an AMC algorithm to maximize communication throughput. For AMC, regular feedback from the receiver to the transmitter is necessary to gather channel state information (CSI). On the one hand, obtaining feedback too often drastically impacts the communication throughput in channels with long propagation delays, but on the other hand, insufficient feedback leads to poor AMC decisions and hence poor throughput. We propose a dynamic feedback strategy to automatically find the right balance and maximize communication performance.