Low Power Complex Multiplication using Pre-computation Technique For FFT Algorithm in Wearable ECG Gadgets

Abstract

The focus on wearable devices for biomedical applications is gaining a lot of research and market interest. Heart diseases remain by far the main cause of death and a challenging problem for biomedical engineers to monitor and analyze. Electrocardiography (ECG) is an essential practice in heart medicine. However, wearable ECG gadgets for real time analysis still faces computational challenges, especially when multiple lead signals are to be analyzed in parallel, in real time, and under increasing sampling frequencies and battery operated gadgets. Another challenge is the computation of Fast Fourier Transform (FFT) on huge amounts of data that may grow to days of recordings. In this research we present the comparative study of FFT calculation best suitable for power optimized applications in the biomedical field and exclusive performance enhancement by reducing dynamic power consumption. ECG application specific to FFT calculation to the final hardware-software (HW/SW) architecture is the focus of this paper.