an area and power efficient signal denoising and morphological feature extraction architectures for ecg signal analysis

Abstract

newline quotElectrocardiogram (ECG) is the analytical form of biological signal that provides important information about the patients. In recent days, it is necessary to use a computer-aided diagnosis system to monitor cardiac patients continually and identify heart illnesses automatically. The recording and transmission techniques frequently affect the morphological structure of ECG records. This distortion affects the accurate detection of disorders related to the cardiac system. Digital system implementations with high performance and low energy consumption continue to be difficult to develop, particularly for portable devices. In real time, the speed of the processors affects the noise reduction and morphological feature extraction process of the ECG data. Field programmable gate array (FPGA) can manage more processes at the same time and it offers an advantage over DSPs in terms of speed because of its parallel nature. newlineIn this research work, a new FPGA design of variable step-size variable tap length delayed error normalized least mean square (VSS-VT-DENLMS) noise removal algorithm is proposed to remove Baseline Wander (BW), Muscle Artifacts (MA), Power Line Interference (PLI) noises from the ECG signal. For the study of ECG signals, a morphological feature extraction method based on FPGA is also suggested. The proposed VSS-VT-DENLMS filter alters the weight update equation of the DENLMS method by simultaneously changing the step sizes and tap lengths. It is used to achieve a better trade-off between rapid convergence and error tracking. In addition, the adaptive filter structure of the proposed VSS-DENLMS is designed by taking into account both systolic and folding structure with compressor-based booth multiplier to improve performance in terms of speed and area, newlineAlso, the primary unit of the proposed morphological feature extraction architecture is the Generalized Synchrosqueezing transform (GSST). A detrended fluctuation analyser is used in the reconstruction stage to capture the most possible info