Biometric Authentication Using Encrypted Vector Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing authentication systems face challenges in handling non-binary data sets with varying degrees of similarity or dissimilarity, leading to reduced security and information leakage, particularly in biometric and password-based authentication methods.
Innovation Solution
A method and system that encodes data into fixed-length vectors, encrypts them, and uses a secure approximate equality operator to determine similarity or dissimilarity without revealing the underlying data, employing techniques like autoencoders, approximate equality operators, and Shamir's secret sharing to preserve privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If biometric identifiers are compared with reduced resolution to achieve repeatability, then authentication reliability is improved, but security is weakened
Solution Approach 1:
The patent segments the biometric comparison process into multiple independent components: extracting local features (minutiae points, ridge characteristics), encoding them separately using binary templates, and comparing only these encoded features rather than full-resolution images. This segmentation allows reliable authentication through feature matching while maintaining security by never exposing or storing complete high-resolution biometric data.
Solution Approach 2:
The patent applies local quality by focusing comparison efforts on specific critical regions of biometric data rather than treating all data uniformly. For fingerprints, only minutiae points and ridge characteristics in specific locations are extracted and compared. This localized approach ensures authentication reliability through precise feature matching while reducing security risks by limiting exposure to only necessary local features.
2Measurement precision
If strict binary comparison is used for data sets with varying similarity, then measurement precision is improved, but information loss increases
Solution Approach 1:
The patent applies preliminary action by pre-processing biometric data into encoded templates before comparison. Features are extracted, encoded into binary representations, and stored in advance. During authentication, only these pre-encoded templates are compared using simple binary operations. This preliminary encoding maintains measurement precision through exact binary matching while preserving similarity information by retaining the essential characteristics in the encoded form.
Solution Approach 2:
The patent creates simplified copies of biometric data in the form of binary templates that capture essential features without containing the original detailed information. These binary copies (e.g., 160-byte fingerprint templates) preserve the necessary authentication characteristics while eliminating redundant data. The copying process maintains precision for authentication purposes while reducing information loss by keeping only the most discriminative features.
3Measurement precision
If complete biometric data is stored and compared, then measurement precision is improved, but information leakage risk increases
Solution Approach 1:
The patent extracts only the essential authentication features from complete biometric data and separates them from the original detailed information. For fingerprints, only minutiae points and ridge characteristics are extracted and encoded into binary templates. The original high-resolution fingerprint images are never stored or transmitted. This extraction maintains measurement precision through accurate feature capture while eliminating privacy leakage risks by removing unnecessary detailed information.
Solution Approach 2:
The patent introduces binary encoded templates as an intermediary between the original biometric data and the comparison process. These templates serve as a mediator that preserves authentication precision while preventing direct access to sensitive biometric information. The intermediary encoding layer ensures that even if templates are compromised, the original biometric data remains protected and cannot be reconstructed from the binary representations.
Data Source
AI summary
According to a method for biometric authentication of a user of a client computing platform, a service-providing system inputs from the client computing platform a candidate vector comprising an encoded, ordered data set that is encrypted using a key generated from a secret entered by the user in the client computing platform. The ordered data set is a digital representation of at least one physical and/or behavioral biometric parameter of the user. Without requiring any knowledge of raw data about the secret or the biometric parameter(s), a comparison value is determined according to a comparison function between the candidate vector and a template vector stored in the service-providing system. When the comparison value meets a predetermined criterion, an authentication message is generated indicating sufficient similarity between the candidate and template vectors.