Design and Evaluation of an Electrodermal Activity Sensor (EDA) With Adaptive Gain

Abstract

Electrodermal activity (EDA) is a physiological signal that can be measured non-intrusively and associated with the sympathetic nervous system, hence being widely used nowadays. Although multiple sensor designs have been proposed over the years, practical hardware limitations still persist, due to differences in the electrode materials used to interface with the body, sweat gland density at the recording sites, and variability in the overall electrical properties of the skin in-between users. This is particularly challenging when low resolution Analog-to-Digital Converters (ADC) are used, which, due to the proliferation of open source and low-cost hardware platforms, are currently widespread. In this article we present a design and experimental evaluation of a new circuit for EDA measurement with adaptive gain control, allowing the dynamic adaptation of the measurement range. The proposed circuit has shown comparable results to those obtained with a reference sensor, obtaining a mean correlation coefficient of 0.967, and proved to have superior performance in cases where the reference sensor would saturate.