Parallel-type fractional zero-phase filtering for ECG signal denoising. (April 2015)

• Record Type: Journal Article
• Authors: Wang, Jianhong; Ye, Yongqiang; Pan, Xiang; Gao, Xudong
• Is Part Of: Biomedical signal processing and control. Volume 18(2015)
• Page Start: 36
• Page End: 41
• Publication Date: 2015-04
• Subjects: Fractional filtering -- Zero-phase filtering -- Parallel-type filter -- Center Grünwald–Letnikov differintegrator -- Electrocardiogram signal denoising; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2014.10.012
• Languages: English
• ISSNs: 1746-8094

Abstract

Abstract : Highlights: The fractional differintegral is used to enhance the compromise capability between ECG signal denoising and ECG signal information retention. The center Grüsnwald–Letnikov differintegral filtering technique is used to obtain zero-phase effect. The proposed technique exhibits excellent performances on visual sense and quantitative comparison. Abstract: In this paper, a parallel-type fractional zero-phase filtering technique based on the center Grünwald–Letnikov differintegrator is proposed. We first present a left and a right Grünwald–Letnikov differintegrators, which are generalized magnitude-and-phase modulations. By using them in parallel we obtain a center Grünwald–Letnikov differintegrator, essentially a parallel-type fractional zero-phase filter. And then a center symmetrical convolution mask is constructed to implement the proposed fractional zero-phase filter. The method presented eliminates the phase distortion while offering a better compromise between signal denoising and signal information retention than conventional filtering methods. To illustrate this, the differintegrator and conventional filters were applied to electrocardiogram signals. The results indicate that the method we propose has superior performance compared with conventional denoising methods.