(A) study on the software-defined sensor interface system

Abstract

The advent of continuous healthcare monitoring in daily life has led to an exponential surge in demand for wearable devices capable of routinely tracking and managing bio-signals. This increase in demand, complicated by the simultaneous need for enhanced performance for proactive health management and small, durable devices, necessitates the evolution of power-efficient sensor interface systems. This study identifies the issues of power wastage inherent in traditional sensor interface circuit design and proposes a system that operates with enhanced power efficiency. This system, utilizing machine learning software for the adaptive control of photoplethysmography (PPG) circuits, is validated through Simulink modeling. By analyzing analog circuits from a system perspective, the study not only improves power efficiency but also introduces novel possibilities in the field of adaptive control of analog systems, leveraging machine learning tools.