Underwater acoustic communications in warm shallow water channels

Abstract

The ability to communicate underwater has applications in maritime commerce, research and defense. As electromagnetic waves cannot propagate well in seawater, acoustic waves are extensively used for communications. However, attempts to communicate acoustically have met with limited success in warm shallow waters. Non-Gaussian ambient noise and time-varying multipath propagation are major performance hurdles. Signal processing techniques based on the Gaussian noise assumption perform poorly in the presence of non-Gaussian noise. Rapid time variation of the channel limits channel equalization. We develop a new statistical model for the ambient noise and a physics-based channel model for the shallow water acoustic channel. We then develop computationally efficient signal processing techniques to provide near-optimal communication performance in the presence of such noise. We show that a coded interleaved orthogonal frequency division multiplexing scheme is well-suited for warm shallow water communications, obtaining better data rates as compared to commercial modems tested in local waters.