LEUVEN, BELGIUMA new research concept uses a combination of body parameters to monitor a person's arousal level. IMEC has developed an ambulatory arousal monitor in a compact form factor with a long battery lifetime. It consumes less power than a comparable Bluetooth device. In fact, the entire system allows several days of autonomy on a commercial prismatic Li-Ion battery.
IMEC's approach within its Human++ program at Holst Centre uses a body-area network that measures four body parameters to detect a person's emotional state. The body-area network consists of two small wireless sensor nodes, which communicate to a PC acting as a base station. The wireless sensor nodes handle ultra-low-power digital signal processing (DSP) and wireless communication of the measured data.
The first wireless sensor node, which is integrated in a chest belt, measures respiration and ECG (electrocardiogram or heart activity) based on IMEC's single-channel biopotential application-specific integrated circuit (ASIC). The second node is integrated in a wristband. It consists of a commercial sensor for skin temperature and a dedicated circuit board measuring the galvanic skin conductance between two fingers. The physiological measurements are combined and interpreted in the software running on the base station. There, an indication about the person's arousal is derived in real time.
Being able to measure and analyze the body's emotional state can be valuable for a variety of applications ranging from medical clinical trials to mobile gaming. In drug screening, for example, being able to objectively quantify parameters like stress can complement the more subjective indication and gradation of traditional tests. Another application is in online gaming, where online avatars can automatically adapt to a player's state of mind without him or her having to actively indicate it in a game menu.
Future research will focus on making the system fully autonomous in several ways. It will use energy harvesters that produce electricity from body heat. In addition, the amount of local computing done within the sensor nodes will be increased along with the amount of parameters that can be measured. Finally, robustness and reliability will be improved including ultra-lowpower wireless RF communication.
See Associated Figure
