Attack Scenarios and Security Analysis of a Blockchain and PUF-based Lightweight Authentication Protocol for Wireless Medical Sensor Networks

Abstract

In Wireless Medical Sensor Networks (WMSNs) wearable or implantable sensors are used to collect vital body parameters, allowing remote monitoring and advanced predictive and preventive healthcare. This involves transmitting patients’ physiological data, which are sensitive and should be confidential, over the network. Thus, ensuring the security and privacy of these data is one of the most important prerequisites for the successful development of healthcare systems based on WMSN. In this regard, reliable authentication protocols designed for this specific scenario are needed. In this paper, we propose the most efficient attacks to date on the widely referenced lightweight authentication protocols for WMSNs - Wang et al.’s protocol (IEEE Internet of Things Journal, 9(11), 2022, doi: 10.1109/JIOT.2021.3117762). This protocol relies on blockchain and smart contracts to address the issue of centralization, and on Physically Unclonable Functions (PUFs) to ensure advanced physical layer security. Our attacks exploit the protocol’s vulnerabilities that stem from an inadequate use of the underlying blockchain component, unlike the previous attacks which require challenging conditions to meet, i.e. physical access to users’ mobile devices and specialized power analysis techniques. We introduce new attack scenarios and prove that Wang et al.’s protocol is vulnerable to impersonation, tracing, message replay and session key disclosure. We validate these vulnerabilities using the ProVerif tool, thereby confirming the feasibility of the attack scenarios and refuting Wang et al.’s claims regarding the protocol’s security properties. In addition, we identify and discuss other design weaknesses in Wang et al.’s proposal that further undermine its overall security and practical applicability. Finally, we revisit the feasibility of previously published attacks on Wang et al.’s protocol.