A ICA applied to passive ocean acoustic tomography

International audienceIn last two decades, many researchers have been involved in acoustic tomography applications. Many algorithms have been developed for active tomography applications. However few algorithms have been dedicated to the passive acoustic tomography applications. Up to now, all the developed algorithms are considering the single input single output channel or some specific type of signals (narrow band with specific features or signatures in time-frequency domain). This paper deals with the blind separation of many natural and artificial acoustic sounds, in order to simplify and improve the performance of the previously cited algorithms. First of all, the choice of some Independent Component Analysis (ICA) algorithms is discussed, then some simulation results are given and analyzed