Ocean sensing using ambient noise: An overview
Karim G. Sabra
Abstract
Karim G. Sabra
Abstract
The random nature of ocean ambient noise, natural or man-made, tends to suggest limited utility: ambient noise is commonly considered to be a nuisance for acoustic sensing or imaging applications. However, whether noise is a nuisance or a useful signal actually depends on how it is processed. Indeed, acoustic fields from random sources are often considered to be incoherent, but there is some coherence between two sensors that receive signals from the same individual noise source. Consequently, by cross-correlating the ambient noise recorded at two hydrophones one can estimate the acoustic waves that propagate between them, thus yielding an estimate of the Green's function between those sensors. This passive approach provides a foundation for ocean sensing and imaging techniques using only ambient noise, thus without active transmitters. We will examine the basic background physics of extracting coherent information from ambient noise. Specifically we will highlight among others the role of frequency/bandwidth, noise structure/distribution, ocean fluctuations, sensor separation and, if employed, the array configuration on this process. Further we will give an overview of recent experimental results in shallow and deep water and discuss practical challenges for implementing noise-based ocean sensing techniques.
A significance statement is not available in the OpenAlex record.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
The random nature of ocean ambient noise, natural or man-made, tends to suggest limited utility: ambient noise is commonly considered to be a nuisance for acoustic sensing or imaging applications. However, whether noise is a nuisance or a useful signal actually depends on how it is processed. Indeed, acoustic fields from random sources are often considered to be incoherent, but there is some coherence between two sensors that receive signals from the same individual noise source. Consequently, by cross-correlating the ambient noise recorded at two hydrophones one can estimate the acoustic waves that propagate between them, thus yielding an estimate of the Green's function between those sensors. This passive approach provides a foundation for ocean sensing and imaging techniques using only ambient noise, thus without active transmitters. We will examine the basic background physics of extracting coherent information from ambient noise. Specifically we will highlight among others the role of frequency/bandwidth, noise structure/distribution, ocean fluctuations, sensor separation and, if employed, the array configuration on this process. Further we will give an overview of recent experimental results in shallow and deep water and discuss practical challenges for implementing noise-based ocean sensing techniques.
Key concepts: Ambient noise level, Acoustics, Noise (video), Noise floor, Computer science, Coherence (philosophical gambling strategy), Seismic noise, Background noise