Covert underwater communication through cepstrum modulation mimicking Pseudorca crassidens whistles using machine...

Area of Science:

• Marine Biology
• Acoustic Communication
• Signal Processing

Background:

• Growing demand for secure underwater acoustic communication.
• Need for covert data transmission methods in marine environments to avoid detection by intruders.
• Limitations of existing methods in ensuring true secrecy.

Purpose of the Study:

• To introduce a novel technique for covert underwater acoustic communication.
• To embed secret information using cepstrum transform to imitate false killer whale whistles.
• To achieve secure communication undetectable by eavesdroppers.

Main Methods:

• Mimicking false killer whale whistles (Pseudorca crassidens) for data embedding.
• Utilizing cepstrum transform for watermark embedding.
• Employing machine learning to assess signal imperceptibility and robustness.
• Conducting underwater experiments for validation.

Main Results:

• Demonstrated exceptional robustness and improved data capacity.
• Achieved a superior bit error rate in underwater experiments.
• Maintained a high watermark capacity with a low probability of recognition.
• Successfully validated the model for secure communication networks.

Conclusions:

• The proposed method offers a novel and effective approach to covert underwater acoustic communication.
• The technique successfully conceals secret information within acoustic signals, mimicking natural whale vocalizations.
• The model shows significant promise for enhancing security in underwater communication systems.