Bio-inspired underwater acoustic communication through PCHIP-based whistle generation and improved CSS modulation

Abstract

Bio-inspired underwater acoustic communication has emerged as a promising approach for secure underwater communication by camouflaging signals as marine mammal vocalizations. However, conventional bio-inspired communication methods predominantly rely on fixed cetacean whistle patterns or limited whistle databases. The inherent predictability in signal frame structures significantly boosts their vulnerability to pattern recognition and adversarial detection. This consequently weakens the fundamental premise of bio-inspired communication systems, namely, their capacity to emulate natural marine mammal vocalizations for covert operations. To address this limitation, our research group proposed a novel bio-inspired secure communication model utilizing a piecewise cubic hermite interpolating polynomial (PCHIP) for simple whistle generation. The proposed approach ensured frequency continuity in signal design while maintaining high similarity to natural cetacean whistles. Subsequently, an improved chirp spread spectrum (CSS) modulation scheme was developed, and a PCHIP-based spread spectrum algorithm was introduced to generate random time-frequency spectral profiles. Particularly, randomly generated synchronization headers were incorporated with sliding correlation in the proposed system, enabling individual whistles to achieve both synchronization and information transmission capabilities. The integrated method significantly reinforced the covertness of bio-inspired communication while maintaining communication efficiency. The simulation results in shallow water environments demonstrated the effectiveness of our approach. Meanwhile, sea trials successfully achieved communication over 10 km distances with a bit error rate of ${10}^{-4}$, validating the practical viability of the proposed system.