Abstract
This paper discusses in detail the extraction methods of the foetal Electrocardiogram (fECG) signal from the non-invasive maternal Electrocardiogram (mECG) and enhancement techniques. Foetal ECG compound signals from DaISy database obtained non-invasively by placing electrodes on the abdomen area of the mother were used for the purpose. In this paper two algorithms were used, Extended State Kalman Filter which uses modified nonlinear effective model and priory generated synthetic ECG signals and FastICA which uses the order of Kurtosis. Finally Hilbert-Huang Transform (HHT) was applied to extracted fECG signal to describe the non-linear and non-stationary nature of signals by instantaneous frequency which reveals the intra-wave frequency modulations. Also the efficiencies of these methods were compared based on few important criterions such as output waveforms, PSNR and energy function of the extracted signal. Based on these findings the best method in extracting foetal ECG from noise contaminated source signals was presented.