Biometric Authentication Segmentation for Large-Scale Identity Verification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric authentication systems face challenges in efficiently authenticating large numbers of enrolled individuals while maintaining a desired security level and user convenience, as the False Acceptance Rate (FAR) and False Rejection Rate (FRR) are interdependent, and increasing the number of characteristic features or affine regions can decrease convenience and increase system costs.
Innovation Solution
A method and system that utilize biometric sensors to record and compare biometric features with prestored data, applying identifying and verifying acceptance tolerances to achieve secure authentication, with the option to select prestored features for verification, and employing anonymous identifiers for secure communication, including visual codes and NFC links for convenient data transfer.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If biometric features are compared with prestored features of large numbers of enrolled persons, then authentication coverage is improved, but system complexity and response time worsen
Solution Approach 1:
The patent segments the authentication process into two distinct phases: identification phase where biometric features are compared against a large database of enrolled persons using loose tolerances, and verification phase where the identified person's prestored features are compared using strict tolerances. This segmentation allows the system to handle large numbers of enrolled persons efficiently by separating the computationally intensive database search from the security-critical verification step.
Solution Approach 2:
The patent performs preliminary identification action by comparing current biometric features with prestored features of multiple enrolled persons before final verification. This preliminary comparison narrows down the search space from millions of enrolled persons to a small candidate set, significantly reducing the computational burden of the subsequent verification phase while maintaining comprehensive authentication coverage.
2Reliability
If acceptance tolerances are lowered to meet high security levels, then False Acceptance Rate is improved, but False Rejection Rate worsens
Solution Approach 1:
The patent dynamically adjusts acceptance tolerances based on the authentication phase. During identification, looser tolerances are applied to enable matching across variations in biometric capture conditions. During verification, stricter tolerances are applied to ensure high security. This dynamic adjustment allows the system to maintain low False Acceptance Rate while minimizing False Rejection Rate by adapting the tolerance level to the specific operational context.
Solution Approach 2:
The patent segments the tolerance application into two phases: identification phase with identifying acceptance tolerances and verification phase with verifying acceptance tolerances. The identifying tolerances are configured to meet response time requirements for large-scale authentication, while the verifying tolerances are configured to meet security level requirements. This segmentation resolves the contradiction by applying appropriate tolerance levels at appropriate stages.
3Reliability
If multiple characteristic biometric features are recorded, then authentication reliability is improved, but convenience and system costs worsen
Solution Approach 1:
The patent applies partial action by recording and comparing only the most discriminative biometric features necessary for reliable authentication, rather than all possible features. The system identifies and utilizes key characteristic features that provide sufficient distinction between individuals, achieving high authentication reliability with a limited set of features, thereby maintaining user convenience and reducing system costs.
4Ease of operation
If biometric authentication is performed without possession verification, then user convenience is improved, but security level worsens
Solution Approach 1:
The patent creates a universal authentication system that can operate in multiple modes: standalone biometric authentication for high-convenience applications, and biometric-plus-possession verification for high-security applications. The system's multi-functionality allows it to adapt to different security requirements while maintaining a consistent technical foundation, enabling organizations to choose the appropriate authentication strength based on their specific needs.
Data Source
AI summary
A method and a system enable authentication of large numbers of enrolled persons. At a biometric sensor, current biometric features of a person are recorded. At a matching server, currently recorded biometric features of the person are compared with prestored biometric features of enrolled persons while applying one or more identifying acceptance tolerances, and positive authentication is signalled in case of a match. At a user device of the person, in case of no positive authentication at the matching server, selecting prestored biometric features of the person at the matching server is enabled for comparison with the currently recorded biometric features while applying a verifying acceptance tolerance, and positive authentication is signalled in case of successful verification.