Multi-factor Authentication in Innovative Business Systems of Industrial Clusters

Abstract

Development of information and communication technologies and the Internet significantly modifies tools and approaches to innovation process implementation. At current stage, the importance of creating continuous mutually conditioning interaction of individual business systems of industrial clusters is emphasised. Rapid progress in innovation infrastructure development, ensuring the implementation of scientific and industrial cooperation, affects almost all stages of this process. Ensuring the data transmission security in such systems is implemented using methods and models of symmetric and asymmetric cryptography, which are currently used in almost all spheres of human activity, including banking, e-mail protection and corporate data. However, the critical vulnerabilities presence in their operation leaves the developing new methods and algorithms problem for ensuring security in the implementation of information and communication exchange between individual industrial entities in clusters still relevant. To solve it within the framework of this study, a multi factor authentication method was developed based on a dynamic model of increasing the target function using a deep learning neural network model. The developed method of multi-factor authentication is based on dynamic patterns prediction acting as keys for two-way automatic authentication. Among the main advantages of this algorithm are increased security due to the use of a key pair on the client and server side, as well as the absence of the need to re-enter the authenticated sequence. An important implemented system feature is the support for the separation of information flows into parallel data protection keys for isolated transmission and processing of information and increased security.