Cavitation noise is produced by the formation and collapse of bubbles on the surface of underwater propellers. Acoustic recordings of cavitation noise are broadband and noise-like, yet distinctly rhythmic as the propeller blades rotate through water flow inhomogeneities. With Prof. Les Atlas (UWEE) and Dr. Ivars Kirsteins (NUWC), our goal is to develop maximum-likelihood demodulation algorithms for characterizing the rhythmic components of cavitation noise. By treating the carrier as a stationary noise signal, the maximum-likelihood approach precisely defines what modulations can and can't be detected from a random process.
- P. Clark, I.P. Kirsteins, and L. Atlas, "Existence and estimation of impropriety in real rhythmic signals," Proc. IEEE ICASSP, Kyoto, Japan: 2012. [pdf]
- P. Clark, I. Kirsteins and L. Atlas, "Multiband analysis for colored amplitude-modulated ship noise," Proc. IEEE ICASSP, Dallas, TX, 2010. [poster (PDF)]
Photographs of cavitating propellers.
Modulation spectrum of actual cavitation data, showing the fundamental propeller rate of 2.8 Hz with varying intensity depending on carrier (acoustic) frequency.