Fuzzy Biometric Query Protocol for Private Real-Time Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric surveillance systems face challenges in maintaining privacy while performing real-time identification, as they often store and process biometric data in plaintext, leading to potential privacy breaches and inefficient search operations that scale linearly with database size.
Innovation Solution
A privacy-preserving fuzzy query system using a fuzzy labeled set intersection (FLPSI) protocol that encrypts and processes biometric data through multiple-party computation, allowing clients to learn only the identity of matching data without revealing the content to the server, and achieving sublinear communication costs as the database grows.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If biometric data is stored and processed in plaintext for real-time identification, then search functionality and identification accuracy are improved, but privacy protection deteriorates and security risks increase
Solution Approach 1:
The patent creates encrypted copies of biometric data that can be searched and processed without revealing the original plaintext information. The system generates cryptographic representations (hashes, encrypted vectors) that preserve the ability to perform identification while preventing unauthorized access to actual biometric data, thus resolving the contradiction between productivity and privacy protection
Solution Approach 2:
The patent introduces cryptographic intermediaries (encryption algorithms, hash functions, secure multi-party computation protocols) that mediate between the need for data processing and the need for privacy protection. These intermediaries allow the system to perform real-time identification on encrypted data without exposing the underlying biometric information, thereby maintaining both productivity and security
2Object-affected harmful factors
If biometric data is encrypted to protect privacy, then security and confidentiality are improved, but search efficiency and computation speed deteriorate
Solution Approach 1:
The patent performs preliminary actions by pre-processing biometric data into encrypted representations (hashes, feature vectors) before actual search operations. This preprocessing creates optimized cryptographic forms that enable faster searching and comparison operations on encrypted data, thus improving search speed while maintaining security protection
Solution Approach 2:
The patent changes the parameter representation of biometric data from plaintext to various encrypted formats (binary vectors, hash values, cryptographic primitives). These parameter transformations enable efficient cryptographic operations and comparisons that are faster than traditional encrypted search methods, resolving the contradiction between security and search efficiency
3Object-affected harmful factors
If traditional private matching protocols are used for fuzzy search, then privacy preservation is improved, but communication and computation resources increase linearly with database size
Solution Approach 1:
The patent segments the biometric data representation into fixed-length feature vectors with specific dimensions (e.g., 128-dimensional binary vectors). This segmentation allows for efficient comparison operations that do not scale linearly with database size, as each segmented feature can be independently processed and compared using optimized cryptographic operations, thus reducing overall resource consumption while maintaining privacy preservation
Solution Approach 2:
The patent implements fuzzy matching that requires only partial correspondence (e.g., t-out-of-T matching criteria) rather than exact matches. This partial action approach reduces the computational burden by allowing approximate matches to be identified more efficiently, decreasing the resources needed for searching large databases while still preserving privacy through cryptographic methods
Data Source
AI summary
An exemplary system and method are disclosed that provide the private querying of a biometric scan, such as a person's photo, against a private database such that the client issuing the private query learns only the identity of the query if it is in the database while the server performing the processing learns nothing of the biometric scan. The exemplary system and method conduct privacy-preserving searches over fuzzy databases via a fuzzy labeled set intersection (FLPSI) protocol. FLPSI protocol computes an intersection of noisy input sets to assess for closeness/similarity metrics in an efficient and optimized search.