This thesis investigates the use of wavelets, wavelet packets and cosine packet signal decompositions for the removal of noise from underwater acoustic signals. Several wavelet-based denoising techniques are presented and their performances compared. Results from the comparisons are used to develop a wavelet-based denoising algorithm suitable for a wide variety of underwater acoustic transients. Performances of the denoising algorithm are compared to those of a short-time Wiener filter implementation, and demonstrate that wavelet-based methods are a viable tool for the denoising of acoustic data.

Back to homepage.