A Novel Trust Management Model in the Social Internet of Things

The Internet of Things (IoT) and social networking integration, create a new concept named Social Internet of Things (SIoT) according to which the things are able to autonomously establish social relationships with regard to the owners. Things in SIoT operate according to a service-oriented architecture. There may be misbehaving owners and consequently misbehaving devices that can perform harmful attacks based on their social relationships with other things for their own gain at the expense of other IoT devices. This motivates us to work on the issue of how to estimate the trust of a service provider to avoid malicious service providers and select the best service provider. In this paper, a novel trust management model is proposed based on four properties. The model deals with attacks (especially on-off attacks) and considers service levels for services provided by each node. A method to provide different levels of services via SIoT devices, and a new trust assessment scheme are the contributions of this paper. We evaluated the proposed scheme with extensive simulations and the results show that the proposed model can effectively select the best service provider and cope with most trust related attacks.