Two Channels Fetal Electrocardiogram Extraction by Combining Continuous Wavelet Transform and higher order statistics via JADE Algorithm

This paper provides an addition to several works that address the problems of blind source separation using only the statistical methods such as ICA and PCA. It deals with the fetal electrocardiogram extraction by combining the continuous wavelet transform CWT and the higher order statistics. In these two last decades, the JADE algorithm has been widely used to solve the problem of separation of the fECG and mECG signals from 8 channels. The present work will show that it is possible to avoid this large number of channels because, in fact, it is not the number of channels that counts but rather the quality of the channel in terms of energy and information that it carries. That is why, before applying the statistical algorithm we suggest making a selection of channels by using the Continuous Wavelet Transform CWT. Adopting this method, the number of channels will be reduced to 2 channels.