Abdominal signals (ADS) recorded from pregnant women represent an important tool for monitoring the fetal heart rate (FHR) variability and the well-being state of the fetus, mainly because it has the advantage of being noninvasive. Thus, no risk is given during recording either for the mother or for the fetus, but complex signal processing steps are necessary, mainly due to the presence of the maternal ECG in the ADS, in order to achieve a clean fetal electrocardiogram (fECG). The paper presents an improved application of the Event Synchronous Canceller (ESC) for maternal electrocardiogram (mECG) suppression. An adaptive mECG template which also includes the P and T waves is considered for the ESC algorithm. ESC is able to perfectly separate the mECG even though the fetal beats overlap with the maternal QRS complex (mQRS). The algorithm is applied both on real ADS recorded during pregnancy and on simulated ADS data; the latter now uses simulation of all signal and noise components, including the fetal and maternal ECG. The modified ESC shows good results in extracting a cleaned fECG signal that can help in reducing the inconsistency in interpretation of FHR. Thus the false positive diagnosis regarding the health status of fetus is strongly reduced.