A Service Oriented Adaptive Trust Evaluation Model for Ubiquitous Computing Environment

The increased participation of various devices and networks, involvement of heterogeneous system, and context-awareness raises a significance of trust and trust evaluation in the ubiqutioius computing environment. Consequently, the trust evaluation based on the context and service requirements is adapted to decide the applicable level of security to the offered services. Therefore, this paper presents a service-oriented adaptive trust evaluation model (SOATEM) in the ubiquitous computing network. The proposed trust evaluation model computes direct trust based on various interaction properties and recommendation trust by filtering the dishonest recommenders based on the time context. A balanced blending of direct and recommendation trust are done with trust weights to make them relevant to current application scenarios. The results are compared with some of the available schemes and found to be a consistently good in performance for the today's ubiquitous network.