Ambient noise in acoustic underwater communication modeling and analysis

Underwater communication of autonomous underwater vehicles (AUVs) has challenging demands on acoustic communication systems caused by the changing signal propagation, ambient noise, multi-path, and other effects in the underwater acoustic channel. Performance tests of new communication systems in the form of sea trials are very costly, while simulations and emulations provide a cost-effective alternative. However, these simulators/emulators require a model of the modem behavior and structure, the underwater channel, the communication process, and the changing environment. In this paper, we present a model-based approach to generate ambient noise as a part of the underwater channel model with finite impulse response (FIR)-filters over the desired power spectral density (PSD). We compared the FIR-filter approach with commonly used alternatives, like additive white Gaussian noise (AWGN) and Knudsen based noise. The results are based on a synthesizable underwater pass-band modem in MATLAB/Simulink with quadrature phase-shift keying (QPSK) modulation. We show that the ambient noise generated by FIR-filters has the formally desired PSD and provides an up to 20% better estimation of the decoding rate for a selectable frequency spectrum compared to AWGN-channel.