Revocable Biometric Tokens via Modular Arithmetic
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Biometric systems face challenges in providing secure and revocable identity verification due to issues with intra-subject variations, sensor variations, and privacy concerns related to the permanent nature of biometric data, which existing transforms and encryption methods fail to adequately address.
Innovation Solution
The implementation of multi-part transforms with robust distance metrics that allow for cryptographically secure biometric data transformation, enabling secure revocable identity tokens that can be matched without decryption, supporting multiple versions and incorporating a second factor for enhanced privacy and security.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional biometric transforms and encryption methods are used, then security is improved, but intra-subject variations and sensor variations cause matching accuracy to deteriorate
Solution Approach 1:
The patent transforms biometric data by changing its mathematical parameters through modular arithmetic operations. The biometric template is transformed using parameters (a, b, m) where the transformed template T' = a*T + b (mod m). This parameter transformation provides security through cryptographic strength while the modular arithmetic structure preserves the distance metric properties needed for accurate matching despite variations.
2Adaptability or versatility
If biometric data is stored permanently for identification, then identification capability is improved, but privacy and security are worsened due to inability to revoke compromised data
Solution Approach 1:
The patent implements dynamic biometric templates that can be updated and revoked. Instead of static permanent storage, the system allows generation of new transformed templates with different parameters (a, b, m) when compromise is detected. The binding structure allows dynamic re-enrollment where old templates are invalidated and new ones created, providing revocability while maintaining identification capability.
Solution Approach 2:
The patent segments the biometric template into multiple components that can be independently transformed and stored. The template is divided into feature vectors that undergo separate modular transformations, allowing selective updates and revocation of specific template portions without requiring complete system re-enrollment.
3Adaptability or versatility
If multiple biometric databases are maintained for different applications, then application versatility is improved, but unauthorized correlation and privacy breaches are worsened
Solution Approach 1:
The patent applies different local transformation parameters (a, b, m) to biometric data depending on the specific application or database context. Each application receives biometric data transformed with application-specific parameters, making the data locally optimized for that application while appearing as unrelated noise in other contexts, thereby preventing unauthorized correlation across databases.
Data Source
AI summary
Techniques, systems and methods relating to cryptographically secure revocable biometric signatures and identification computed with robust distance metrics are described. Various biometric cryptographically secure revocable transformation approaches are described that support a robust pseudo-distance computation in encoded form, thereby supporting confidence in verification, and which can provide for verification without identification.