Dolphin whistle-inspired underwater acoustic signal recognition via refined composite multiscale amplitude-sensitive permutation entropy

Dolphin whistle-inspired underwater acoustic communication signals exhibit significant camouflage capabilities, complicating their discrimination not only from natural dolphin whistles but also among themselves, owing to the high similarity among different biomimetic modulation types. This inherent similarity renders traditional features ineffective for accurately identifying specific biomimetic modulation schemes. To address this challenge, this paper analyzes the characteristics of dolphin whistles, Bio-LFM, and Bio-Cos signals, and proposes a novel feature extraction algorithm based on the refined composite multiscale amplitude-sensitive permutation entropy (RCMS-ASPE). Furthermore, a complete recognition framework is established to handle the simultaneous interference from dolphin whistles and bionic signals, achieving effective recognition of dolphin whistles, Bio-LFM, and Bio-Cos signals. Simulation results demonstrate the recognition performance of the proposed method for Bio-LFM and Bio-Cos signals and reveal the relationship between the recognition rate and modulation parameters. To further validate the proposed recognition method, a sea trial was conducted in the South China Sea, demonstrating its effectiveness in real-world oceanic conditions. • Robust recognition scheme for dolphin whistle-inspired underwater acoustic signals. • Novel RCMS-ASPE algorithm enhances feature extraction. • Links recognition rate to signal modulation parameters. • Aids biomimetic communications & countermeasures development.