Computational trust models for collaborative network orchestration

Simple attacks can be countered by simple technical measures, but a defence against organized attacks requires collaboration amongst service providers. In order to create effective defense strategies, sharing cyber intelligence amongst service providers is therefore becoming increasingly important. Additionally, networks have grown in scale, complexity, and degree of inter-connectedness, such that their protection can often only be guaranteed and financed as a shared effort. To create a collaborative network amongst different service providers to facilitate the sharing of information and cyber threat intelligence we need to organize, maintain and evaluate trust amongst the autonomous members who have their own desire and goals to achieve that may result in conflicting interests.
The series of studies reported upon in this thesis were conducted in the context of the SARNET project (Security Autonomous Response networks). The SARNET project aims to provide an ICT system that can defend itself autonomously. To design such a system, the SARNET project looks at three different layers, the Strategic, Tactical and Operational layer. This thesis considers questions at the Strategic layer, where we studied what is needed to create and maintain a cyber security alliance. At this level, this research focuses on the question “What dynamic computational trust models enable cyber-intelligence sharing through partner selection for collaborative cyber defense operations?”. In this research, we operationalized the trust concept and researched the impact of trustworthiness factors on the success of a collaboration.