Regularized second order source separation

摘要: The task of separating signals from experimentally measured linear mixtures is often complicated by the presence of noise sensor noise and statistical dependencies between the original sources, which often makes standard independent component analysis (ICA) algorithms fail [1, 2]. One way to overcome these problems is to introduce additional knowledge we have about the mixing process and the signals themselves.Here we suggest to add a regularization term to the cost function of multishift extended spatial decorrelation (multishift ESD,[2]) which contains prior information about the time-course of one or more original sources. Using an artificial toy dataset and a dataset that contains prototype signals obtained from optical recording of brain activity we show that the regularization term improves the separation results at different noise levels.