Multi-Biometric Authentication to Reduce Spoofing and False Positives
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional single-biometric authentication systems are susceptible to false positives and spoofing attempts, failing to provide adequate security and accuracy in user verification, particularly in insecure environments.
Innovation Solution
Combining two or more biometric identifiers, such as facial recognition and voice recognition, enhances authentication security and accuracy by reducing the likelihood of false positives and spoofing, optionally integrated with smart contract functionality.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If single-biometric authentication is used, then user experience and convenience are improved, but security and accuracy deteriorate due to false positives and spoofing susceptibility
Solution Approach 1:
The patent combines multiple biometric identifiers (e.g., facial recognition, voice recognition, gait analysis) into a unified authentication system. By merging multiple independent biometric modalities, the system achieves both high security through multi-factor verification and improved user experience through streamlined authentication flow, resolving the contradiction between convenience and security.
Solution Approach 2:
The system creates a composite authentication approach by integrating different biometric types (physiological and behavioral characteristics) into a unified verification framework. This composite approach leverages the complementary strengths of different biometrics to achieve both accuracy and usability, preventing false positives while maintaining ease of operation.
2Device complexity
If single-biometric authentication is used, then device complexity is reduced, but measurement precision and accuracy deteriorate
Solution Approach 1:
The patent segments the authentication process into multiple independent biometric verification stages. Each biometric identifier (facial, voice, gait) is processed separately through dedicated recognition modules, then the results are synthesized. This segmentation allows high precision through multiple verification stages while managing complexity through modular architecture.
Solution Approach 2:
The system transitions from single-dimension authentication to multi-dimensional verification by incorporating multiple biometric modalities. This dimensional expansion increases measurement precision and accuracy by verifying identity across multiple independent characteristics, while the patent manages the resulting complexity through efficient data fusion techniques.
3Reliability
If multiple biometric identifiers are combined, then authentication accuracy and security are improved, but device complexity and computational requirements worsen
Solution Approach 1:
The patent performs preliminary actions by pre-processing and extracting features from multiple biometric identifiers before the actual authentication decision. Feature extraction, normalization, and preliminary verification are performed in advance to reduce the computational burden during the final authentication phase, thereby achieving high robustness while managing system complexity.
Solution Approach 2:
The system introduces an intermediary data fusion module that mediates between multiple biometric verification results and the final authentication decision. This intermediary layer synthesizes information from different biometric sources in an efficient manner, achieving enhanced authentication robustness while controlling computational complexity through optimized fusion algorithms.
Data Source
AI summary
Systems and techniques may be used for multiple biometric data authentication. An example technique may include receiving first biometric data and second biometric data, optionally combining the first biometric data and the second biometric data into combined biometric data, and comparing the first and second or the combined biometric data to stored biometric reference data associated with an account of a user to generate a similarity score. The example technique may include determining whether the similarity score exceeds a threshold value, and in response to determining that the similarity score exceeds the threshold value, outputting an indication.