Extraction of FECG Signal Based on Blind Source Separation Using Principal Component Analysis Mahesh B. Dembrani, K.B. Khanchandani and Anita Zurani Abstract Fetal electrocardiogram (FECG) gives faithful medical information of heartbeat rate of the fetal living. Extraction of FECG from abdomen of maternal woman consists of interferences and motion artifacts and noises. Maternal elec- trocardiogram (MECG) is a main source of interference signal present in FECG. This paper focuses on FECG extraction from blind adaptive filtering using principal component analysis (PCA). The abdominal ECG (AECG) is obtained by blind adaptive algorithm which consists of MECG and FECG QRS complex. Principal component analysis separates the two MECG and FECG. The experiments show that it can simultaneously accomplish maternal ECG and fetal QRS complexes enhancement for their detection. The simulation results show that FECG extracted from the peaks of R-R interval is noise-free signal, and extract FHR. Keywords FECG extraction ⋅ MECG ⋅ Blind source separation (BSS) ⋅ Principal component analysis (PCA) 1 Introduction Blind source separation is referred as BSS, and the term blind says that the signal obtained is the linear mixture source that are independence sources. The signal consisting of mixture sources can be analyzed by number of different techniques. Blind source separation based on principal components analysis (PCA) and inde- pendent component analysis (ICA) can be used to analyze such mixture signals [1]. M.B. Dembrani ( ✉ ) ⋅ K.B. Khanchandani ⋅ A. Zurani Shri Sant Gajanan Maharaj College of Engineering, Shegaon 444203, Maharashtra, India e-mail: mahesh.dembrani@gmail.com K.B. Khanchandani e-mail: kbkhanchandni@rediffmail.com A. Zurani e-mail: anita.zurani@gmail.com © Springer Nature Singapore Pte Ltd. 2018 P.K. Sa et al. (eds.), Progress in Intelligent Computing Techniques: Theory, Practice, and Applications, Advances in Intelligent Systems and Computing 518, DOI 10.1007/978-981-10-3373-5_17 173