Practical consideration for ICA

Independent component analysis (ICA) and blind source separation (BSS) refer to the problem of recovering statistically independent signals from a linear mixture. And its success for a given data set may depend crucially on performing application-dependent preprocessing steps. Before applying ICA some of the things we have to be careful about concerning many parts of ICA. Of course there are more things one has to keep in mind in order to apply ICA successfully, or at least with the biggest chance to be.