Search. Read. Cite.

Easy to search. Easy to read. Easy to cite with credible sources.

Asian Journal of Information Technology

Year: 2016  |  Volume: 15  |  Issue: 6  |  Page No.: 1049 - 1055

A Multi-Stage Adaptive Singular Value Decomposition Approach for Fetal ECG Signal Extraction in Multichannel Input System for Prenatal Health Monitoring

Ravindrakumar Selvaraj and Bommannaraja Kanagaraj


A novel methodology for denoising and extraction of Fetal ECG signal is presented. The study is extended for the classification of the fetal ECG. A hybrid multistage adaptive filtering with singular value decomposition technique to remove the noise and to extract the Fetal Electrocardiogram signal from the abdominal signal is proposed. The proposed architecture is a combination of Singular Value Decomposition (SVD) method and Adaptive extraction method using QR-decomposition and affine projection algorithm. This method leads to enhancement of fetal ECG by canceling maternal ECG and noises in the first stage using SVD. The proposed method is applied to simulate and real time ECG signals to demonstrate its superior effectiveness. The Proposed methods are superior in its performance when compared to existing methods like independent component analysis, wavelet transform and others.