Authenticating tier-two body area network devices through user-specific signal propagation characteristics

Wireless body area network (WEAN) plays a vital role in patient health monitoring due to its ability to measure private physiological data of patients via low-power sensor devices and send them to medical experts through wireless channels. However, the open nature of wireless channels makes the collected patients' data vulnerable to unauthorized access. Thus, effective authentication and data privacy protection in WEAN are important. The devices in WEAN lack adequate resources to handle existing complex cryptographic authentication solutions. Consequently, non-cryptographic based authentication approaches are proposed, which are more desirable because of their less complexity. However, the existing methods limit themselves to securing intra-WEAN communications, leaving the data of the patients unprotected against security attacks in the inter-WEAN layer. Moreover, they either require additional sensors, specified user movements, or restricted proximity, which undermines their widespread adoption. To address these shortcomings, we propose a hybrid device authentication scheme based on a user-specific signal propagation signature to secure inter-WEAN communications using channel characteristics and lightweight symmetric encryption. The key ideas in the proposed scheme are constructing signal propagation signatures related to each user from received signal strength (RSS) data and leveraging a feedforward neural network to correctly recognize the signal propagation signatures of the users and use them to verify inter-WEAN devices positioned on the bodies of the users. We conduct security and performance analyses of our scheme to show its resistance to security attacks and its computational effectiveness. Moreover, we implement our scheme on hardware devices and conduct extensive experiments on human subjects in indoor and outdoor places to validate the robustness of the proposed method. The results of the experiments demonstrate that our approach can effectively detect 92.7% of attack attempts with false alarms on just 7.5% of legitimate traffic. (C) 2022 Elsevier Ltd. All rights reserved.

Automated summary

Wireless body area network (WEAN) plays a vital role in patient health monitoring, but its openness makes patient data vulnerable to unauthorized access. Existing authentication methods limit their security to intra-WEAN communications and fail to protect patient data in the inter-WEAN layer. To address this, we propose a hybrid device authentication scheme based on user-specific signal propagation signatures and lightweight symmetric encryption to secure inter-WEAN communications.