Fundamental Numerical Analysis of Biomimetic Pulse Train Signals for Underwater Acoustic Ranging: Proposal of a Pulse Train Generation Method

Abstract

A biomimetic pulse train signal is proposed for underwater acoustic ranging in autonomous underwater vehicles (AUVs). The pulse train mimics the searching phase of dolphins as they emit short pulses continuously at approximately constant intervals and amplitudes. To enhance biological realism, fluctuations in interval timing and amplitude are introduced into each pulse. These variations also provide advantages for signal processing. Accurate travel time estimation is crucial for acoustic ranging. The proposed 127-pulse train, with a nominal inter-click interval of 29 ms, incorporates temporal fluctuations of up to 10⁻² s and amplitude variations within 25%. It exhibits a sharp autocorrelation main peak and low sidelobes, and its correlation properties are superior to those of M-sequence signals of the same length. By mimicking natural dolphin click trains, the signal reduces the likelihood of introducing conspicuous anthropogenic cues. The signal also has a cumulative sound exposure level of 184.8 dB re 1 µPa²s, below auditory injury thresholds for high-frequency cetaceans. This approach offers a promising option for covert AUV operations and potential applications as a pulse compression signal in underwater acoustic systems.