Improving security levels in Automatic Teller Machines (ATM) using multifactor authentication

Thumbnail Image
September, 2016
Journal Title
Journal ISSN
Volume Title
A wide variety of systems need reliable personal recognition systems to either authorize or determine the identity of an individual demanding their services. The goal of such systems is to warrant that the rendered services are accessed only by a genuine user and no one else. In the absence of robust personal recognition schemes, these systems are vulnerable to the deceits of an impostor. The ATM has suffered a lot over the years against PIN theft and other associated ATM frauds. In this research is proposed a fingerprint and PIN based authentication arrangement to enhance the security and safety of the ATM and its users. The proposed system demonstrates a three tier design structure. The first tier is the verification module, which concentrates on the enrollment phase, enhancement phase, feature extraction and matching of the fingerprints. The second tier is the database end which acts as a storehouse for storing the fingerprints of all ATM users’ preregistered as templates. The last tier presents a system platform to relate banking transactions such as balance enquiries, mini statement and withdrawal. The system is developed to run on Microsoft windows Xp or higher and all systems with .NET framework employing C# programming language, Microsoft Visio studio 2010 and SQL server 2008. The simulated results showed 96% accuracy, the simulation overlooked the absence of a cash tray. The findings of this research will be meaningful to the Bank of Ghana (BoG) and the Ghana Association of Bankers (GAB).
A Dissertation Presented to the Kwame Nkrumah University of Science and Technology, Faculty of Physical Sciences, in Partial Fulfilment of the Requirements for the Award of the Master of Science in Information Technology,