Identification of individuals for telemetric personal health monitoring systems

Abstract

Prophylactic measures are reducing costs of healthcare without lowering its quality. Telemetric Personal Health Monitoring (TPHM) Systems with multi-parametric sensors allow an active integration of a patient into his treatment.

A common problem in such a system is the verification and identification of the user to guarantee the authenticity of the gathered data.

To assign the measured physiological parameters to the legitimate user we included a gait-based identification algorithm into a TPHM System.

The algorithm records two-dimensional acceleration data from a two-axis accelerometer located on the chest belt. In the configuration mode the participant has to walk fast for one minute to allow the algorithm to calculate a reference pattern for the legitimate user of the belt. In the identification mode the algorithm is looking for fast walking sequences and compares the respective pattern with the reference pattern. Test patterns generated by 4 users wearing the belt were compared to find out if the algorithm is able to distinguish between legitimate and illegitimate users.

Analog acceleration data are sampled at 250Hz for 20 seconds and converted into a two dimensional acceleration diagram. The reference pattern is generated from the three most typical patterns out of five runs by the legitimate user. We use histogram statistics as a simple and performance-efficient way to classify the acquired data, as our goal is to implement our algorithm directly on the chest-belt’s microcontroller.