A novel method for estimating source number of fetal ECG

Fetal ECG (FECG) is an important method of monitoring the health of the fetus, dropping perinatal morbidity and mortality, and it has been a hot topic recently, so many methods of extraction of fetal ECG are proposed. At the same time, blind separation algorithms are feasible and effective for the extraction of fetal ECG, but these algorithms are normally assumed that source numbers are known and there is only a fetal signal mixed with the maternal signal. In fact, a pregnant woman may have multiple births, and the source number is unknown, so the previous blind separation algorithms for the extraction of fetal ECG are no longer applicable. This paper proposes a novel method to estimate the number of fetal ECG based on improved fuzzy C-means (FCM) clustering, by which the numbers of fetal ECG and mother ECG can be estimated. Thus the novel algorithm provides the priori knowledge for the subsequent blind extraction method of FECG. Finally, the simulation results show that the method is effective and accurate.