Impact of sea waves on performance of shallow water acoustic communications

摘要

Shallow water acoustic channels have fast time-variance, long-time multipath spread and large Doppler shift. Their transfer characteristics are influenced by many factors, such as: the operating frequency; the acoustic characteristics of sea surface and bottom; the ocean sound speed profile; the water depth; the depth and distance between the transmitters and the receivers; the variations of channel boundary; and, subsea objects. A time-varying channel impulse response has both deterministic and stochastic characteristics in a shallow water environment. Given the ocean environmental parameters, the deterministic impulse response can be calculated exactly by using underwater sound propagation models. However, it is extremely difficult to predict the stochastic impulse response due to its complexity. The method of ray tracing is quicker and more efficient than other methods such as with normal modes, parabolic equation and wave number integration in order to predict underwater water acoustic channels. At high frequencies, the ray tracing method is as accurate as others. Therefore, the ray tracing method has been widely used to simulate the channel impulse response in underwater acoustic communications. Here we focus on the impact of sea waves on the performance of single-carrier coherent underwater acoustic communications. Assuming that the sea wave is a single sinusoidal wave, we modify the BELLHOP ray module included in the Acoustic Toolbox in order to calculate the time-varying channel impulse response in a shallow water environment. Four time-varying channel impulse responses are presented for sea waves with a wave length of 50.0 m, and wave heights of 0.0, 0.5, 1.0 and 2.0 m. Furthermore, we investigate the impact of the wave height on the performance of the single-carrier coherent underwater acoustic communication system. Simulations demonstrate that bit error rates (BERs) of the system remain unchanged with time when the wave height equals 0.0 m. However, BERs change rapidly with time when the wave height is greater than 0.0 m. The higher the wave height, the faster the channel impulse response changes with time, and as a result, the higher the mean BER.