Trust Metrics Based on a Trusted Network Element

In this paper we study and propose a trust model and trust metric composition models based on a trusted network element. Our proposed model, executed in a trusted network element, helps the user to make subjective decisions based on secure and trusted metric information presented in a userfriendly form. The composition models present two possible solutions of how to integrate trust constructs into quantitative measurements in order to provide readily available evidence to the trustor about trustee's trustworthiness. The results show how to achieve 5% measurement error probability when detecting malicious actions and what kinds of 95% intervals of confidence the 5% measurement error probability will enable in different trust metric composition models. The presented trust metric is especifically designed for client-server and peer-to-peer communication scenarios over the Internet, such as Web browsing and/or content streaming.