Trustworthy of Implantable Medical Devices using ECG Biometric

Abstract

Implantable medical devices (IMD) such as pace-makers, and cardiac defibrillators are becoming increasingly interconnected to networks for remote patient monitoring. However, networked devices are vulnerable to external attacks that could allow adversaries to gain unauthorized access to devices/data and break patient privacy. To design a lightweight computational trustworthy of IMD, we propose novel ECG-based biometric authentication using lift and shift method based on post-processing data from the noise generated in an ECG signal recording. The lift and shift method is an ideal addition to this system because it is a quick, lightweight process that produces enough random bits for encrypted communication. ECG is a signal that is already being measured by the IMD, so this ECG biometric could utilize the data that is already being actively recorded. We provide a comprehensive evaluation across multiple NIST tests, as well as ENT and Dieharder statistical suites test.