This research by Tamir Mishali, Paolo Casari, and Roee Diamant focuses on developing a method to differentiate between authentic underwater sounds made by marine animals and artificial signals that mimic these sounds. This distinction is crucial for enhancing the security of underwater communications, especially as the "Internet of Underwater Things" continues to grow. Underwater Acoustic Communication (UAC) is widely used for applications such as oceanographic data collection, coordination between scuba divers, and operations of underwater vehicles.

The methodology involves analyzing the CIR of sounds using two metrics: cross-correlation, which measures how similar two CIRs are, and sample entropy, which assesses the randomness in the CIRs. Two interception measures are then applied. The first is the Kullback-Leibler Divergence (KLD), which compares the distribution of CIR similarities to a Gaussian (normal) distribution. A closer match to a Gaussian distribution indicates more random, animal-like sounds. The second measure is clustering, which determines how many distinct groups (clusters) the CIR similarities form. Fewer clusters suggest a more stable, biomimicking source.

Results from both simulations and real-world experiments demonstrate the effectiveness of this method. The research tested the approach with both artificial signals and actual dolphin recordings, successfully distinguishing between stable (biomimicking) and unstable (animal) sound sources. Field trials conducted in a lake environment further validated the method's ability to accurately classify sounds under varying conditions.

In conclusion, this research presents a robust approach to intercepting and classifying underwater acoustic signals. By focusing on the stability of the CIR, it provides a reliable means to distinguish real animal sounds from artificial, biomimicking ones, thereby enhancing the security of underwater communications. This work contributes significantly to the advancement of covert communication techniques in underwater acoustic environments and lays the foundation for further exploration of interception methods in this domain.

For the full article, you are welcome to visit:

https://ieeexplore.ieee.org/document/10529203

By Tamir Mishali