DESIGN AND EVALUATION OF A SECURE BIOMETRIC ATM USING FINGERPRINT RECOGNITION

Abstract

The increasing prevalence of fraud and security breaches in Automated Teller Machine (ATM) systems has highlighted the limitations of traditional card-and-PIN-based authentication methods. This study presents the design, implementation, and evaluation of a fingerprint-based biometric ATM system aimed at enhancing transaction security and user authentication. The proposed system replaces conventional knowledge-based authentication with identity-based verification using fingerprint recognition, thereby reducing vulnerabilities associated with credential theft, card skimming, and unauthorized access.

The system architecture integrates biometric sensing, embedded processing, and secure communication mechanisms to ensure reliable and efficient operation. A minutiae-based fingerprint recognition algorithm is employed for feature extraction and template matching, while secure encryption techniques are used to protect sensitive biometric and transaction data. The system was implemented using a microcontroller-based platform and evaluated under controlled experimental conditions.

Performance evaluation was conducted using standard biometric metrics, including authentication accuracy, False Acceptance Rate (FAR), False Rejection Rate (FRR), and response time. The results indicate an authentication accuracy of 96.5%, a FAR of 2.1%, an FRR of 4.3%, and an average response time of 1.8 seconds, demonstrating the system’s effectiveness in balancing security and usability. Functional and security testing further confirmed the robustness of the system, achieving a high success rate across all test scenarios. The findings suggest that fingerprint-based biometric authentication provides a secure, reliable, and user-friendly alternative to traditional ATM systems. The study concludes that the proposed system is suitable for real-world deployment in modern banking environments, with potential for further enhancement through multi-factor authentication and advanced biometric techniques.