Abstract
Long-term ambient noise recordings contain an immense amount of information which quickly becomes cumbersome to analyze fully and efficiently. Correlation matrices provide a promising approach to identifying and isolating the active source mechanisms in ambient noise datasets. By comparing the noise levels in different frequency bands, frequency regions which tend to change together are identified, pointing toward either one common source or multiple related sources for the identified frequency bands. An attempt at simplifying the frequency and temporal information provided by series of correlation matrices is presented, utilizing data recorded by the Comprehensive Nuclear-Test Ban Treaty Organization’s hydroacoustic monitoring system.
Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
