Application of wavelet transforms to sonar signal detection and estimation

The conventional role of the sonar processor was to isolate long duration tonal signals corrupted by underwater environment noise. With the trend towards acoustic signature reduction in submarine operation, traditional sonar processing techniques are proving inadequate. This reduction in the available signal power has fostered an interest in both alternative spectral processing methods and noise reduction techniques. The properties of the Wavelet Transform (WT) make it a good mathematical tool for multiresolution decomposition of signals. The main thrust of this thesis is to investigate the possibility of obtaining detection and estimation improvements using the WT. This thesis focuses on the use of the Discrete Wavelet Transform (DWT) to display underwater acoustic spectral information and enhance spectral display using noise reduction techniques.;Using simulation and data collected from a passive sonobouy receiver, the WT is shown to provide an acceptable representation of the spectral information. The flexibility afforded by the WT and the ease and speed by which the display resolution can be modified are demonstrated. The performance of noise reduction functions is shown to enhance both the lofargram and the WT presentation of signal information.