UNIVCHAIN: um modelo para autenticação de documentos acadêmicos baseado em blockchain

Abstract

Amid constant technological advances, academic document validation remains a challenge for universities, students, and the labor market. Documents commonly issued on paper have some disadvantages in terms of an efficient way of checking, storing, and free distribution to third parties. Verification of these documents usually occurs through manual labor of the issuer and involves costs. The paper is a fragile material and can be useless in a short time under adverse conditions. The use of these documents for academic proof can only be done by sending physical or digital copies, which invalidates much of the security resources applied in the document. Also, these do not always have all the relevant information of the student and his course. This work presents a model based on blockchain that allows the issued and the verification of academic documents without the need of intermediaries or to resort to the issuing entity to attest their veracity. The proposed model makes use of public blockchain that allows broad ac- cess, applicability, and cost reduction. The data model is based on an open standard, facilitating adoption, extension, and improvements. The use of an open model also benefits the internation- alization of Brazilian graduates. The standard of digital signatures proposed by ICP-Brasil, and used in this work, makes the documents issued by the model, legally valid in Brazil. The characteristics of this model make it unique when compared to the models found in the literature, providing three security checks on the documents issued. The Digital Signature is the first check, followed by the fingerprint verification at blockchain and lastly the revocation check. The application of this model can be made concomitantly with the physical emission already carried out today by the academic entities. The results were collected through the construction of a prototype based on the proposed model and the realization of use cases based on a real case. The prototype showed little variation in the emission times, where 75% of the documents issued took up to 422 seconds. The most significant variation perceived was identified in the block’s confirmation time, a step that took three times as long as the other steps added up. Besides, the prototype demonstrated a low demand for memory resources.