Selecting the preferred biometric authentication method

Abstract

In this article, a multicriteria methodology is used to compare some biometric authentication methods with other identification methods. In order to analyze biometric access control systems qualitatively, it is necessary to take into account the characteristics of the systems, as well as to apply certain methods for multicriteria optimization. Therefore, it is necessary to conduct a comparative analysis of biometric authentication methods, taking into account a set of quality indicators. This paper compares certain biometric access control systems by a set of quality indicators, namely, the false-admission rate (FAR), which means a certain percentage when a situation arises when the system can allow access to a user who is not registered in the system, and the false-rejection rate (FRR), which is the percentage of access denied to a real user of the system. Both parameters are calculated using mathematical statistics. The lower the value of these indicators, the higher the quality of authentication of certain users. The article compares and uses the hierarchy analysis method to select the preferred option of known biometric identification tools based on the analysis of fingerprint, 2D human recognition, iris, retina, vein pattern, keyboard handwriting and voice. The hierarchy analysis method is based on the decomposition of the task of selecting the desired design option for a particular system and dividing this task into simpler parts, using the estimates obtained from experts for pairwise comparisons of simple parts of the task. After processing certain data, the preferred biometric identification method is determined. The use of biometrics can solve reliability problems and can improve the reliability of authentication and identification of certain objects when accessing a system with a large number of users, as well as very important and necessary systems as part of physical access control and management systems.