From Personal Experience to Global Reputation for Trust Evaluation in the Social Internet of Things

Abstract

Abstract-Trust has been exploring in the era of Internet of Things (IoT) as an extension of the traditional triad of security, privacy and reliability for offering secure, reliable and seamless communications and services. It plays a crucial role in supporting IoT entities to reduce possible risks before making decisions. However, despite a large amount of trust-related research in IoT, a prevailing trust evaluation model has been still debatable and under development. In this article, we clarify the concept of trust in the Social Internet of Things (SIoT) ecosystems and propose a comprehensive trust model called REK that incorporates third-party opinions, experience and direct observation as the three Trust Indicators. As the convergence of the IoT and social network, the SIoT enables any types of entities (physical devices, smart agents and services) to establish their own social networks based on their owners relationships. We leverage this characteristic for inaugurating Experience and Reputation, which are originally two concepts from social networks, as the two paramount indicators for trust. The Experience and Reputation are characterized and modeled using mathematical analysis along with simulation experiments and analytical results. We believe our contributions offer better understandings of trust models and evaluation mechanisms in the SIoT environment, particularly the two Experience and Reputation models. This paper also opens important trust-related research directions in near future.

Trust Evaluation Mechanism for User Recruitment in Mobile Crowd-Sensing in the Internet of Things

“…We therefore use the second method in this simulation and, with the damping factor set to 0.85, the error tolerance = 10 3 , and the number of users ranging from 200 to 1000, it takes from 25 to 32 iterations to converge. This reputation calculation is suitable for huge networks like the IoT as the scaling factor is roughly linear in logarithm of N [7].…”

Strengthening the Blockchain-Based Internet of Value with Trust

“…Therefore the Reputation defined in Equation (8) is also existed and unique. Details of the mathematical proofs and analysis for the Reputation model can be found in [22].…”

Quality of Information as an indicator of Trust in the Internet of Things

“…To establish and evaluate trust relationships between service requesters (i.e., trustors) and service providers (i.e., trustees), we leverage the REK trust model in the IoT proposed in [5,10], which consists of the three major trust indicators, namely Reputation, Experience and Knowledge. Knowledge indicator is as "direct trust" inferred from attributes of a trustee whereas Experience and Reputation are as "indirect trust" calculated from previous interactions illustrating personal opinion and global perspective toward the trustee, respectively.…”

“…Knowledge indicator is evaluated as QoI score; other trust-related attributes are neglected due to unavailability or not being suitable to collect; however, in some use-cases, some other useful information could be gathered and plays as supplemental factor in evaluating Knowledge. The two indicators Reputation and Experience are calculated based on previous research works [5,10] and briefly presented below.…”

“…Then, we propose a way to generate a score for these dimensions. We finally evaluated our results by utilising these scores as an important indicator of trust called Knowledge in the Reputation-Experience-Knowledge (REK) trust model proposed in [10,5] for evaluating trustworthiness of IoT services and applications.…”