Bio Acoustic Sources Segmentation in a Noisy Underwater Environment Using Wavelet Denoising with Ica

An advanced type of Independent Component Analysis (ICA) algorithm is used in this paper for separation of biological sounds in underwater noisy environment. Underwater noisy environment are formed by echolocation clicks, snapping shrimps and dolphin whistles mixing. Different types of ICA algorithms like Infomax, Fast-ICA are used to apply on mixture of bio-signals. Depend upon separability performance, effectiveness of these algorithms has been compared. Using ICALAB and open source, signals are available and ICALAB simulation packages are based on this to compare the algorithms of ICA. Finally, graph for proposed technique is shown for comparisons using wavelet denoising.