On Designing Context-Aware Trust Model and Service Delegation for Social Internet of Things

Abstract

Social Internet of Things (SIoT) an emerging Internet-of-Things (IoT) service infrastructure, which integrates the social concept into IoT systems for enhancing service efficiency. By establishing a social relationship among objects, IoT devices can autonomously interact with each other free from human intervention. One critical issue in the development of SIoT is the trust issue, which is essential in fostering cooperation among objects. The existing approaches largely fall short in trust quantification, and also suffer from the bias of trustworthiness evaluation without properly considering inherently dynamic context changes and potential malicious behaviors. In this work, we combine social trust theory and incorporate the unique characteristics of IoT devices to address the trust issue in SIoT. We establish a general trust model which comprehensively captures the competence, willingness, and social relationship in SIoT. Specifically, we define two functions in terms of the Degree of Importance (DoI) and the Degree of Contribution (DoC) to compute the competence and willingness, and we present a formal quantitative trust model that is robust in the dynamic environment and against common malicious attacks. The effectiveness of our proposed trust model is verified through the security analysis and a series of simulation experiments. The results demonstrate that the proposed trust model is reliable and efficient in promoting the success rate of services, as well as improving the efficiency and security of services in SIoT.