Encrypted Biometric Liveness Checks Against Spoofed Authentication
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional biometric systems are vulnerable to faked or replayed biometric signals, lacking effective measures to ensure liveness and privacy in authentication processes.
Innovation Solution
Implementing a biometric authentication system that incorporates liveness detection through randomized biometric requests, using encrypted feature vectors and deep neural networks to analyze and validate biometric inputs, ensuring that the biometric data is contemporaneous and genuine.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional biometric systems are used for authentication, then the authentication process is simple and fast, but the system is vulnerable to faked or replayed biometric signals
Solution Approach 1:
The system performs preliminary liveness detection by analyzing biometric signals for characteristics of genuine live subjects before completing authentication. This preliminary check prevents faked or replayed signals from proceeding to the final authentication decision, thereby improving security without requiring complete system redesign
Solution Approach 2:
The patent introduces an intermediary liveness detection layer between the biometric signal input and the authentication decision. This intermediary component analyzes intermediate features of the biometric signal to determine if it originates from a live subject, adding a security checkpoint without significantly increasing overall system complexity
2Reliability
If liveness detection is added to biometric systems, then security against spoofing is improved, but the authentication process becomes more complex and time-consuming
Solution Approach 1:
The system merges liveness detection analysis with the primary biometric authentication process by analyzing the same biometric signal for both authentication and liveness characteristics simultaneously. This consolidation allows security verification without requiring separate authentication and liveness checks, thereby minimizing additional time requirements
Solution Approach 2:
The patent implements partial liveness detection by focusing analysis on specific key characteristics of genuine biometric signals rather than examining all possible features. This selective approach provides sufficient liveness verification to prevent spoofing while avoiding the time cost of comprehensive analysis of every signal attribute
3Reliability
If randomized biometric requests are implemented, then protection against pre-recorded gestures is improved, but the system complexity and processing requirements increase
Solution Approach 1:
The system implements dynamic randomized requests that change during the authentication process, requiring the subject to perform unpredictable biometric actions. This dynamic approach prevents use of pre-recorded gestures while the randomization is generated and managed through software logic rather than complex hardware mechanisms, controlling the increase in system complexity
4Measurement precision
If multiple biometric types are processed, then identification accuracy is increased, but the system complexity and computational requirements increase
Solution Approach 1:
The patent implements a universal biometric processing framework that can handle multiple biometric types (facial recognition, voice, gestures) through a common architecture. This multi-functional system processes different biometric modalities using the same liveness detection and authentication logic, increasing identification accuracy while avoiding the complexity of separate dedicated systems for each biometric type
Data Source
AI summary
In one embodiment, a set of feature vectors can be derived from any biometric data, and then using a deep neural network (“DNN”) on those one-way homomorphic encryptions (i.e., each biometrics' feature vector) an authentication system can determine matches or execute searches on encrypted data. Each biometrics' feature vector can then be stored and/or used in conjunction with respective classifications, for use in subsequent comparisons without fear of compromising the original biometric data. In various embodiments, the original biometric data is discarded responsive to generating the encrypted values. In another embodiment, the homomorphic encryption enables computations and comparisons on cypher text without decryption of the encrypted feature vectors. Security of such privacy enable biometrics can be increased by implementing an assurance factor (e.g., liveness) to establish a submitted biometric has not been spoofed or faked.