Although trust is viewed as an important attribute of current DRE systems, there is no clear consensus on its definition. A majority of the prevalent approaches also consider trust from an external view, which prevents an integration of trust concepts when developing individual software services and ensembles created out of them. This research therefore is developing a generic model for trust that considers both the internal and external views of trust and an associated formalism based on theory of evidence, composition patterns and the wp-calculus. Such a model will assist the DRE system developers with identifying and measuring trustworthy aspects during early phases of the software lifecycle and can significantly reduce the associated development costs. More importantly, since the research is focusing on fundamental aspects of developing trustworthy DRE systems, the concepts and lessons learned will serve as a foundation for future of Distributed Systems.

Thus, there is a need for an effective and efficient framework, that is flexible and cost effective, for composing trustworthy distributed software.

The proposed approach involves:

  • Extensive survey of existing approaches of modeling trust
  • A preliminary model of trust which extend the subjective logic based preliminary (b, d, u) trust model to incorporate various QoS parameters
  • Guidelines for specifying and verifying the quality of components and component complexes (results of compositions of components),
  • Design a trusted discovery service selection and composition framework which perform an empirical validation of the enhanced trust model

This research proposes to develop a comprehensive framework for designing and implementing trustworthy distributed real-time and embedded (DRE) systems using principles of trust models, subjective logic, multi-level specifications and matching, theory of evidence, and continuous testing.