Wearable biosensors and mobile healthcare (mHealth) technologies are revolutionizing modern healthcare delivery by providing access to pragmatic, medical-grade services that are scalable outside of the hospital setting. Electrodermal activity (EDA) is a physiological signal of particular importance within the mHealth community because it is a useful marker for the physiological arousal of the sympathetic nervous system used in studies of depression, anxiety disorders, stress management, and many more. EDA refers to electrical variations in skin conductance and capacitance occurring at the surface of the skin due to changes in sweat secretion. Modern EDA biosensors often require significant analog and digital resources to acquire and record high-quality EDA signals due to the wide range and variability of skin conductivity across populations. There are significant challenges in maintaining a balance between high-performance sensing capabilities of a biosensor and its ability to be small in size, unobtrusive, and long-lasting. The work within this thesis addresses the research question of how low-resource digital design can be used to improve the size, power efficiency, and utility of wearable EDA sensors while maintaining high-quality physiological sensing capabilities. An ultra-low resource system for EDA measurement is presented that implements a quasi-digital EDA sensor topology for measuring both the

conductive and capacitive components of the EDA signal and requires no analog-to-digital converters or in-phase and quadrature demodulation. Additionally, we apply on-board compression and storage of the EDA signal within a 16-bit microcontroller to improve sensor size and power efficiency by removing the external data storage and transmission requirements for long-term EDA monitoring. The accuracy, precision, dynamic range, and power efficiency of the developed system is characterized and the devices are evaluated in a pilot study.