With rapid developments of network technologies, database outsourcing is emerging as an important new trend beside the "application-as-a-service". In this model, data owners ship their data to external service providers. Service providers do data management tasks and offer their clients a mechanism to manipulate outsourced databases. Since a service provider is not always fully trusted, security and privacy of outsourced data are significant issues. These problems are referred to as data confidentiality, user privacy, data privacy and query assurance. Among them, query assurance takes a crucial role to the success of the database outsourcing model. To the best of our knowledge, however, query assurance, especially for outsourced XML databases, has not been concerned reasonably in any previous work.

In this paper, we propose a novel index structure, Nested Merkle B+-Tree, combining the advantages of B+-tree and Merkle Hash Tree to completely deal with three issues of query assurance known as correctness, completeness and freshness in dynamic outsourced XML databases. Experimental results with real datasets prove the effeciency of our proposed solution.