The Internet of Things (IoT) is a paradigm where entities or things are interconnected, often in heterogeneous contexts. As the interconnection happens, things establish collaborations with others, sometimes under uncertainty. Although trust can help us overcome this uncertainty, things might not be able to process the information about trust coming from other things: each thing could have its own trust model, which means its own way to understand and measure trust. If new trust relationships are to be established, it would be desirable to have a mechanism of interoperability that allows the things to process the information about the other things in terms of trust. In this paper, we describe an interoperability framework for tackling the trust interoperability issues in IoT, depending on the different types of trust models that might co-exist in the same IoT scenario.

Introduction

The Internet of Things (IoT) is based on the interconnection of heterogeneous entities (things) with different connection capabilities and made by different manufacturers [1]. At present, this paradigm is being used in real life applications such as smart homes, health, drones or smart parking [2], [3], [4], [5].

The strength of IoT scenarios lies on one of their main features, that is, the interaction among things. This means that things have to work together towards a common goal. For example, in a smart home environment a smart fridge would have to interact with a smart cook and both with a smart watch or smart phone. Besides the interactions, the heterogeneity of the entities or things involved in these scenarios is another key feature of these environments [6].

In a setting as the IoT, where the interconnection of entities is a fact, the interactions among them have to be a reality in order to make as much profit as possible [7]. The advantages of IoT could be jeopardized if users do not perceive its goodness or even worse, if they perceive their use as a menace for their security or privacy [8]. One of the security problems that arise in these scenarios is that the interactions have to happen even though in most of the cases the behavior of the other entity involved in the interaction is uncertain, that is, the entities involved might not know how the others are going to perform when working towards their common goal. It is then when trust management systems could be of help as they overcome the lack of certainty that traditional security mechanisms are not able to solve. Trust management systems ease the decision-making process by helping the entities involved in any interaction to make informed decisions concerning their relationship with others [9]. In an IoT setting, interoperability has been always an issue [10]. Moreover, if the entities involved are supported by trust mechanisms the most likely will be that each of them have their own trust model, which means their own way to understand and measure trust. Thus, it could be the case when different trust managements systems using different languages and ways to obtain trust have to interact among them. It is then when we need mechanisms that guide the interaction between trust management systems, i.e., we need the trust management systems to be interoperable. This way, the trust information that is derived by any entity could be used and understood by another one, if needed.

Conclusion

The amount of objects or things that are interconnected in IoT systems makes sometimes difficult to handle interoperability issues when referring to trust models for IoT. In this paper, we have presented a framework to deal with interoperability of trust issues. The main idea behind the framework is to consider an Interoperability Trust Repository that deals with the main interoperability facets of IoT such as semantic or syntactic interoperability issues. Moreover, we have performed the validation of the proposed framework with a use case scenario and we have shown that different trust models in different things can interact obtaining common trust values.

In the future, we intend to apply this framework together with the suggested implementation approach to more real IoT use case scenarios such as smart home scenarios. We will also concentrate on the quantification of trust for the different types of models. It will also be interesting to investigate the use of our framework in other decentralized systems different than IoT where the entities involved in it have to interact and use different trust models. It would also be of interest to investigate how the framework performs when small devices are clustered into larger virtual IoT nodes. Moreover, in the case where there are multi-trusted domains.