Recognition method for the bionic camouflage cetacean whistle modulated by CPMFSK signals

Abstract

Bionic covert camouflage underwater acoustic communication has drawn great interests in recent years as a promising way for covert underwater acoustic communication. However, there is little relevant literature to research the recognition of bionic camouflage communication signals (BCCS). Concentrating on the countermeasures for the BCCS, this paper proposes a recognition method for a bionic cetacean whistle signal modulated by continuous phase multiple frequency shift keying (CPMFSK) signal. Based on the features that CPMFSK modulated whistle signals carry the information by imitating the tremble phenomenon of the real whistles, a method is designed to extract a tremble signal which vibrates among the time–frequency contour (TFC) of cetacean whistles. As the tremble signal of the real cetacean whistle and bionic whistle has obvious time, and frequency distinction, focusing on these distinctions, characteristics like normalized power spectrum mean, wiener entropy, etc. are extracted as the characteristic vectors for SVM classification to achieve recognition. Simulations demonstrate the accuracy of the recognition method in different conditions, including modulation parameters, SNR, and underwater acoustic channels.