Increasing the Reliability of User Authentication Based on Protected Electronic Key and Behavioral Biometrics

Abstract

The rapid spread of software applications in all spheres of human activity necessitates the need to ensure the protection of the data contained in them. Therefore, currently, the issues related to the improvement of authentication methods to prevent unauthorized access to software resources are relevant. To solve these problems, it is important to consider that different types of systems have their own unique requirements for authentication subsystems. In addition, the active development of computer technology makes it possible to break authentication algorithms, which were considered reliable a few years ago. In this regard, the work proposes to improve the reliability of user authentication based on a protected electronic key using the AES algorithm and analysis of the entropy of mouse movements, which is an individual characteristic for each person. The implementation of the electronic key is due to its advantages, such as reduced costs, increasing the convenience of use, reducing the probability of loss or theft, and reducing the risk of attacks through an intermediary. Analyzing the entropy of mouse movements will enable to perform the user authentication based on biometric behavioral characteristics that are universal, unique and permanent. Describing the algorithm of the mouse entropy analysis process, it is worth noting that the feature of tracking dynamic behavioral characteristics is that the user moves the mouse cursor along a complex curve with a speed that changes during the movement. The shape of the curved line and the speed of the cursor movement are determined by a number of physiological and psychological factors, in particular, such as: the size and weight of the hand, the position of the hand and the whole body, the state of the nervous system, user habits, etc. To improve the user authentication algorithm in the protected application, an electronic key using the JSON Web Tokens technology, which allows you to avoid authentication errors, increase the performance and scalability of the application, has been developed. Testing of the developed application will be performed on the basis of unit tests and the study of FAR and FRR indices.