(A) history-based trust computation scheme for spontaneously encountering IoT devices

Abstract

Omnipresence of IoT devices in our daily lives introduces a new threat, trust, in an IoT environment. Many researches have been conducted on the trust management for IoT for last several years. However, they are mostly depending on the direct interaction experience with specific IoT devices, which makes it impossi-ble to apply their schemes when two entities have not met each other. This new relationship is common espe-cially when a user firstly visits a place. In this thesis, a trust management scheme is proposed, which com-putes a user’s trust value of an IoT device in an unvisited place considering both the personal trust from the interaction history and non-personal trust from the general public. By leveraging the interaction history with devices of the same type in the similar situation, the trust value can be computed even without the direct in-teraction history. The evaluation results evidence that the proposed scheme can successfully estimate the user’s trust value of the encountered device without the direct interaction history. In addition, the estimated trust value becomes the more accurate, as the more personal experience is leveraged.