Fusion of Electrocardiogram and Arterial Blood Pressure Signals for Authentication in Wearable Medical Systems

Wearable medical systems allow their users to be monitored continuously without being tethered. They are very useful for tracking patient health deterioration in hospital ER and in patient general wards. It is therefore essential to identify who the data is being collected from. In this paper we present an authentication approach that fuses characteristics of electrocardiogram (ECG) with arterial blood pressure (ABP) to authenticate users. The idea behind the use of multiple physiological signals is that it allows us to ensure that the authentication approach works effectively irrespective of the current state of the user’s health. This is an important requirement given that wearable medical systems might be worn by user who have ailments that causes their physiological signals used in authentication to be “non-standard”. An evaluation of our approach showed that it was over 97% accurate with a false positive rate (e.g., accepting illegitimate users) of 1.2% in identifying the user on whom the system is deployed. Further, it enabled authentication after just 3 seconds of signal measurement.