Algorithm Research For Extracting Fetal ECG

It is important to obtain the pure fetal ECG (FECG) and to trace problems in its heart activity. The adaptive filtering may be under the condition of the signal unknown its statistics characteristics, approach gradually the optimum filtering result and adapt to the change of the random environment of outside through self-adaptation. LMS algorithm is one of the wasted used adaptive filtering. And it was applied the system of the adaptive noise canceling, thus it realized the abstraction of the FECG. Independent Component Analysis (ICA) is such a new algorithm that came out with the development of Blind Source Separation in the past ten years.At the beginning, the thesis summed up the current researches in areas of FECG and LMS, and then introduced the definition, standard mathematic model, hypothetical conditions and four estimation principles of ICA. The thesis also pointed out that ICA can recovery the source signals from the observed signals by certain learning rules. In order to extract the purer fetal ECG accurately, NLMS and two kinds of ICA algorithms, the ICA based on Infomax and fast fixed point ICA (Fast ICA), were introduced in this article and realized through simulation on computer. The result of simulation showed that Fast ICA avoid the unconvergence problem which may appear in the methods of the ICA based on Infomax and NLMS for bad Teaming rate, because Fast ICA needn't choose the learning rate.They were applied to multichannel ECG recording obtained from a pregnant woman. It was observed that the three algorithms were able to extract the FECG. But, the experiment outcomes demonstrate the best performance is Fast-ICA algorithm in this important biomedical application. This method has been proved to be able to extract the desired fetal heart beat signal from the composite abdominal signal precisely by simulative and clinical experiment.