An Energy Aware VLSI Framework for ECG Features Extraction for Wearable Devices in Healthcare

Patients in distant places and those who have just had surgery may benefit significantly from long-term heart monitoring and evaluation made possible by advances in wireless technology. A few variables extracted from a wavelet framework might be used as a rough approximation of the Electrocardiogram (ECG) signals. Because of this, a compressive sensing method for ECG signal compression may be proposed and explored. The ECG shows that there is an overlap among cardiac pulses. This repetition suggests substantial shared physiologic guidance among successive pulse rates. In this study, hybridized form of curvelet transform and rapid integration of independent component analysis (HCTRI-ICA) is proposed to extract the features of the ECG signals. To demonstrate the effectiveness of the proposed framework, the study employs the measures of compression ratio (CR), percentage root mean difference (PRD), mean square error (MSE), and peak signal-to-noise ratio (PSNR) to identify ECG signals. The concepts of parallelism and pipelining may form the basis of a design optimized for maximum throughput and performance. The methodology for determining the curvelet coefficients is straightforward and utilizes a cascading structure. By dividing the Read-only memory (ROM) database, the required amount of storage space can be reduced. Verilog HDL is used to simulate the design's specifications and functionality. Virtex-II Pro FPGA is employed with resource-conserving modeling and synthesizing techniques.