Fortifying Artificial Pancreas: A Hardware-in-the-Loop Simulation Study of Cyber-Attacks and Defense Mechanisms

Artificial Pancreas (AP) is a recent technology developed to aid type-1 diabetic patients in maintaining their recommended blood glucose levels. It mimics the function of the natural pancreas by continuously monitoring the patient’s blood sugar level and delivering an adequate amount of insulin to their body. AP is a closed-looped system involving several device components communicating over wireless protocols. Anomalous behavior of any device component or communication protocol due to a cyber-attack can be life-threatening. Hence, it is essential to make the AP system resilient against cyber-attacks. This paper presents the development of a hardware-in-the-loop (HIL) simulation of AP using the Bergman Minimal Model (BMM), which is then used to simulate the responses of two common but potentially catastrophic (to the patients’ health) cyber-attacks, viz., template attack and denial-of-service (DoS) attack. The paper also presents the development of detection and mitigation algorithms against these attacks.