A Novel Model for Abdominal Electrocardiography of a Pregnant Woman

Abstract

This paper formulizes the stationary abdominal electrocardiography (AECG) from pregnant woman as an instantaneous linear model. Then signal to noise ratio (SNR) parameters are defined based on the model and it supports toward evaluating the performance of different algorithm on extracting fetal electrocardiography (FECG). Quantification noise parameters are also considered in the model. This model is tailored for applying independent component analysis (ICA) to separate FECG and maternal ECG signal from AECG. Based on this model, signal to noise ratio (SNR) can be calculated for simulated signals. It is concluded that ICA works magnificent in order to extract FECG even in SNR = -200 dB using simulated data without quantification noise. It showed that the performance was drastically decreased in existence of quantification noise.