Comparison of denoising techniques for Underwater Acoustic Signals

Abstract: Underwater communication (UWC) is a low frequency communication. Its main application is transfer of data in between underwater instruments such as hydrophones, radars etc. The use of low frequency signal comes from the fact that the electromagnetic signals get absorbed under the water due to their high frequency. UWC is very defiant to work with because of its low frequency and random noise present in its conduit. The main purpose of this paper is to detect and denoise the various underwater acoustic signals using two different techniques i.e. EMD (Empirical Mode Decomposition) and Wavelet Transform. These methods denoise the signal by taking into account the effect of AWGN noise. Different types of threshold techniques have been used to denoise the signals. The simulation results shows that there is significant improvement in SNR of the signal when wavelet-soft thresholding technique is used. It is also observed that with the help of thresholding techniques, noise has been downgraded to certain level.

Keywords: Denoising, Underwater, acoustics, audio, DWT, EMD, Thresholding.