Trustworthy data provisioning for IoT-Based circadian health evaluation systems

Abstract

IoT healthcare systems use the data provided by IoT devices to make automated health decisions and to provide recommendations to users for better health. If the data sent by the device is not trustworthy, it will lead to incorrect health decisions and recommendations. In this paper, we propose a trust computational model for provisioning trusted data to the health index evaluation system (HIES). This HIES uses the trusted data selected by the trust model to calculate the health index (e.g., circadian health index) of a particular user. The objective of the proposed work is to provide a trustworthy system for better circadian health decision making by determining the trustworthiness of the data provided by health IoT devices. Unlike existing trust management protocols, our trust model considers identity-based schema such as device identity and user identity, device health, user feedback, data consensus, and acceptance rate for trustworthy data selection. Our trust computational model is resistant to malicious devices with false identities and identifies the devices that report false data. We present the performance analysis of our trusted data selection model and compare it to the baseline scheme.