Accurate assessment of fetal well-being is one of the most important tasks for obstetricians. It is still difficult to measure fetal electrocardiogram (ECG) during fetal movements. Recently, a new method, blind source separation with reference signals, was proposed for stable measurements. This method distinguishes weak signals from noisy mixed signals with little information about the sources. The aim of this study is to estimate the ability of this method for fetal ECG monitoring and to establish standard fetal ECG electrocardiogram values of normal singletons including during fetal movement. The subjects enrolled were 167 pregnant women with normal single pregnancy from 18-to 41-week gestation, who regularly visited Tohoku University Hospital, and 12 pregnant women with fetal abnormality. Fetal signals were successfully separated in 163 of 179 subjects at 91.1% success rate regardless of fetal movements. Time intervals of ECG (P, PR and QRS intervals and QTc) were measured. The standard curves of each interval through the gestational period were obtained. The data in active phase were compared to that in rest phase and the data obtained from normal and abnormal fetuses were investigated. PR intervals in the rest phase were prolonged compared to those in the active phase. Fetal ECG showed anomalous values such as PR interval or QTc prolongation in the abnormal fetuses. The fetal ECG was measured by the new method with or without fetal movements, and the standard fetal ECG values have been established. This study provides a foundation for further detailed clinical studies.