Specifying a security policy: a case study
CSFW '96 Proceedings of the 9th IEEE workshop on Computer Security Foundations
A logical approach to multilevel security of probabilistic systems
Distributed Computing
A systematic literature review of inference strategies
International Journal of Information and Computer Security
| Hi-index | 0.00 |
In reasoning about policy based on information flow control, we can adopt two different points of view depending on whether we deal with explicit permissions or explicit prohibitions. In both cases, we use an epistemic and deontic logic to formally define information a subject is permitted to know. Then, we show how the causality property [2] can be derived from the explicit permissions point of view and how the non-interference [5] and non-deducibility [11] properties can be derived from the explicit prohibition point of view. However, we argue that the prohibitions enforced by non-interference or non-deducibility are generally too rigid and leads to too strong security properties. On the other hand, the causality property only handlesinternal information flow controls and it must be completed to ensure that the security policy is consistently defined. Hence, the consistency problem is discussed; we proposea general definition and practical conditions toverify that a security policy is consistent. We think that the policy consistency problem is closely related to the so-called inference problem in multilevel databases.