The current technological environment with the increasing growth of critical infrastructures relationships represents a hard challenge for security solutions. In order to achieve trust among devices or other networked agents of a critical infrastructure, the definition of completely new methods is required. TITAN aims to cover this ground.
In fact, it is daring to pursue absolute security for a system, while it is also expensive to implement such a solution. Instead of this approach, trust level assurance is presented as a more natural way to face up to the problem. In that sense, this project will try to provide a rational model for trust, which could be incorporated to current solutions in order to be possible to implement this new approach for security.
The main foreseen research challenges in TITAN are summarized as follows:
- Building trust assurance methodology: The TITAN targeted methodology will provide required steps covering end-to end trust estimation applied to a critical infrastructure, from the observed system modelling till the trust presentation and monitoring.
- Distributed Policy Elaboration: TITAN considers the observed infrastructure in a dynamic context which means that policies (at all levels) will be in the core of the trust estimation processes.
- Multi-agent based Reputation mechanisms: Reputation mechanisms are needed to construct a reputation model to be exploited in the trust estimation
- Aggregation techniques and strategies: Trust values will be estimated based on the knowledge (reputation based) and the experience (learning mechanism) of the agents. Aggregation techniques are needed in order to obtain accurate trust values.
- To define an operational trust model in multi-entities, intra-entity and end-to-end communications.
- To propose the technological and methodological solutions for the implementation of such model.