Abstract: Extraction of fetal electrocardiogram (FECG) is currently focused as a prominent issue of medical research. Theoretically FECG can be extracted from the maternal abdominal recordings by conventional blind source separation algorithm. But in the complexity of the real environment, there must be with the proviso that the recordings need to meet the linear mixing model of Blind Source Separation (BSS). This problem will be more difficult in noisy environment. In view of this fact, our paper introduces a discriminant method of BSS based on FECG correlations analysis. The sparse areas of signals are found firstly, and then we can calculate the correlation values of these areas as a judging criterion on the basis of whether the recordings meet the model of BSS so that FECG can be extracted from these observations by conventional BSS methods. Finally, using this discriminant method, the researchers can obtain better performance of extraction or choose other methods which are more interesting.

Key words: fetal electrocardiogram; correlations; blind source separation; FastICA; comb filter