• GSR
• HRV
• physiological sensors
• Wearable sensor

Nowadays, more and more, wearable devices are required in numerous application fields. The ground gained by this technology ranges from daily activity monitoring to medical applications and social communication.
In such a context, physiological sensors housing devices are inserted, among the most important requisites for these objects there is their miniaturization.
The present thesis work is aimed to design, to characterise and to validate a novel ring-shaped biosensor prototype. In particular, the system is required to monitor galvanic skin response and heart rate variability.
Two commercial solutions have been implemented for this purpose: The Maxim 30100 photoplethysmography and the Seeed GSR sensors. A wearable prototype system, governed by a core ST’s STM32F401RE MCU and equipped with Bluetooth communication module, was obtained. A signal acquisition interface and signal elaboration algorithms have been developed in this work as well.
The system evaluation has been carried out by means of comparison with other commercial devices chosen as gold standard, i.e. the Shimmer’s GSR+ unit and the BioHarness’ Zephyr BH3.0 chest belt.
In order to evaluate the system performances, tests on 6 subjects were carried out. Attention was paid to realize a protocol capable of exploit these sensors in both mental and physical stress activities.
The main GSR and HRV parameters were extracted and the error between the waveforms obtained by our system and the gold standards were compared.
Reliable performances were obtained, especially in those cases in which the subject does not need to perform ample arm movements whilst wearing the system. In particular, the relax and the mental stress phases gave interbeat interval curves errors with a mean linear regression coefficient of determination of 0.7 ± 0.07 and a mean RMSE of 23.9 ± 12.0ms (in a range of 400ms), whilst for the GSR curves a R2 of 0.5 ± 0.3 and a RMSE of 0.0213 ± 0.0124µS (in a range of 0.2333µS) were obtained.
Importantly, the results of this work pave the way to the development of an easy to use wearable sensorised system suitable for health care, driving assisting and activity recognition.