WhoAmIOn: A Technique To Determine Whether All Devices Are Being Used By The Same Person

Wearable devices, with a plethora of sensors, can measure physical and physiological signals from various on-body locations, and log the information against a device owner’s profile. However, when another person uses the same wearable device, the other person’s activities and health-related information are also logged into the owner’s profile, causing mis-logging. One approach to prevent mis-logging is by enabling the owner’s wearable devices to communicate amongst themselves to determine if the same person is using them. In this work, we also explore the possibility of using heart rate and inertial sensor data from multiple devices to determine their co-existence with the same user. We developed the WhoAmIOn system that uses machine learning techniques (as compared to correlation approaches that are currently used) to determine whether the same person is using the devices. Overall, we observed that with the inertial sensor, we achieved an accuracy of 52%, while with a heart rate sensor the accuracy increases substantially to 85% in determining whether the same user is wearing or using the devices at the same time instant in a freeliving setting. This demonstrates that, it is indeed possible to use heart rate sensors to determine if devices are on the same body.