In this paper, we provide a formal approach to model SecureDIS security policies enforced on the execution of a query.
Integrating and enforcing these policies is the responsibility of the mediator during the execution of a query placed by a data consumer.
The current state of art shows that during the execution of a query in a spatial database management system (SDBMS), the query optimizer creates all possible query evaluation plans.
This phase can be very short (execution of a query and immediate return of the query result) or can be long lasting; in any case, while the service is being executed the VSN Manager takes care of its execution.
We have used mature, commercially available products from a single vendor and developed a custom middleware Decision Query Module that identifies the best option for most efficient execution of a query.
The integration location should only grant the execution of a query if the purpose of the query specified by the data consumer matches one of the purposes for which the data items have been collected [13].
Therefore, the integration location should only grant the execution of a query if the security level assigned to the data consumer is higher or equal to the sensitivity label assigned to data items returned by that query.
Therefore, the security policy should grant the execution of a query only if the trust level of the data consumer is equal or higher to the one specified by the data sources' security policies.
However, even when the mediator enforces security policies on the execution of a query that discloses data only to authorised consumers and only if the purpose of the query matches the purpose for which the data has been collected, data leakage threats can still materialise in a DIS.
The use of ASTs can significantly reduce the execution time of a query, often by orders of magnitude, which is particularly significant for databases with sizes in the terabyte to petabyte range, whose queries are designed by business intelligence tools or decision support systems.
