Research and Simulation of FECG Signal Blind Separation Algorithm Based on Gradient Method

Independent Component Analysis (ICA) is a new developed signal separation and digital analysis technology in recent years. ICA has widely used because it does not need to know the signal prior information, which has became the hot spot in signal processing field research. In this study, we firstly introduce the principle, meaning and blind source separation algorithm based on the gradient. By using the traditional natural gradient algorithm and Equi-variant Adaptive Source Separation via Independent (EASI) blind separation algorithm, mixing ECG signals with noises had been separated effectively into the Maternal Electrocardiograph (MECG) signal, Fetal Electrocardiograph (FECG) signal and noise signal. The algorithm separation test showed that EASI algorithm can better separate the fetal ECG signal and because the gradient algorithm is a kind of online algorithm, which can be used for clinical fetal ECG signal of the real-time detection with important practical value and research significance.