Uma abordagem baseada em DLT para garantia da fixidez de repositórios digitais confiáveis

Abstract

The increased ease provides by records in digital format also brings drawbacks, translated on the challenges of ensure the integrity and authenticity of these records continuously, knowing this format is very susceptible to flaws in any layers of one computational system. International organizations and standards draws recomendations about fixity to guide the Trustworthy Digital Repositories (TDRs) in yours missions of ensure the integrity and authenticity of the records kept there for the time it was defined (or indefinitely). The use of cryptography hash functions enables verify the data integrity and yours authenticity then. However, to trust in the integrity verified by a hash function also is needed trust in the hash value of reference, that eventually are susceptible to the same threats of whose data refers. Starting from this worry, this work proposes an approach in which must be made a chaining in Merkle Trees of the hashes from TDR's inumerous digital objects and the resulting root values may be registered in blockchain. The use of Merkle Trees enables the generation of a unique value that links the integrity of a large number of records and the register in blockchain ensure a reliable witness value over time for use in proof-of-timestamp, proof-of-existence and integrity check and audit of TDRs' digital heritage, enabling to ensure their fixity. This work also documents an architecture proposal to a platform for recording and auditing the fixity of digital objects, in which it provides the registration of information related to the digital objects fixity using redundant hash algorithms and blockchain networks and which still allows the record of changes in the fixity of these objects, arising from changes considered legitim on them and foreseen by the Open Archival Information System (OAIS), a reference to the treatment of digital objects in TDRs. Based on this architecture, a version of this platform was developed as a proof of concept with the main resources necessary to demonstrate the application of the proposal, enabling the performance of experiments in a simulated TDR environment, that are also documented here and which translate into practical terms some of the concerns documented in this work and the application of the platform.