Title: Ideas for a Trust Indicator in the Internet of Things
Abstract: The Internet of Things will connect many different devices. In order to realise this, users must be willing to trust the devices and communication that happens automatically. We explore the different meanings of trust and strategies that can be used to determine if something is trustworthy and propose a model for trust that takes into account people, devices, and their connections. The model uses ` a priori andposteriori trust to give an indication of how much a user can trust or distrust the information provided by things. This trust indicator can inform users' decisions on whether or not to use a device or service. The Internet of Things (IoT) refers to uniquely identifiable objects (things) and their virtual representations in an Internet- like structure. This term was first used in 1999 by Ashton (1). Other definitions of IoT have appeared as technology progresses. A thing is a real or virtual object, e.g., a device or a web service, offering one or more services. Since im- plementations of the IoT integrate many things belonging to different actors used for various purposes, we must question to what degree we can trust these things both as a individual entities and as a federation of entities. Even when reading the value from one isolated thing we can identify challenges regarding trust. For instance, when looking at a watch we need to consider whether we can trust the time it displays: the watch might show the wrong time for some reason. We also might be observed looking at our watch, which might breach our privacy since observers might consider us impatient or bored. This is in line with Watzlawick's first axiom that suggests that it is impossible not to communicate (2); as a consequence, humans always reveal some information about themselves. These problems are compounded when extending trust issues to the IoT where we are dealing with many different things. It would, therefore, be helpful to have a mechanism to measure how and when to trust things on the IoT and indicate this to the user; users could then take countermeasures if needed. These measures can be based on an applicable model and theory of trust. This would allow users of IoT services to consider their actions and reduce the costs of checking trustworthiness through other frameworks. Our work is inspired by the use of the IoT in health care where monitoring systems using various sensors and devices form a communication internetwork. These devices also need to communicate with a health care infrastructure. In health care systems, the requirements for privacy, security, integrity and availability are especially high. We have previously analysed the security model in patient monitoring systems (3) and suggested a framework for implementing sensor networks in health care (4). We intend to extend this work with a sustainable trust model. We posit that trust in the IoT is not transparent enough for the user. Our contribution consists of ideas for a trust model and metrics for the IoT including both channels, things, humans, and services. In order to reduce threats regarding security, privacy, and functionality when using the IoT, we distinguish between dimensions ofpriori and ` a posteriori trust, as well as computational, technical, and behavioural trust. The developed trust model and its metrics will be used to a) give indications at to which security mechanisms to employ in the implementations of systems; b) give indications to the users at any time of how much they can trust a system, allowing users the opportunity to consider using devices in the IoT, or to use different parameters or settings; and c) annotate data retrieved from a system with trust values derived from the trust model. In this paper, we first define trust and trust models in Section II before presenting our model and metrics of trust in Section III. Finally, we provide an outlook in Section IV.
Publication Year: 2012
Publication Date: 2012-05-27
Language: en
Type: article
Access and Citation
Cited By Count: 25
AI Researcher Chatbot
Get quick answers to your questions about the article from our AI researcher chatbot