Brown A., Deeks J.

Underwater acoustic sensor arrays are affected by self noise from a large number of different sources. Effective remedial action requires the identification of the relative importance of these various sources. Spectra derived from Fourier transforms of single element time series or beam time series have been found to be inadequate in separating the sources of noise. Transforming the data to the wavenumber domain offers one method of data presentation that can solve this problem. This technique provides a graphical representation of acoustic data known as k-Omega plots or k-space plots. The paper explains the generation methodology and illustrates the interpretation of the plot. This method is applied to an example of underwater acoustic data from an array of piezo-electric sensors. The paper further discusses the benefits of the method within the analysis task and some limitations.