Encrypted Text Matching via Auxiliary Data and Hamming Distance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing biometric authentication methods fail to adequately protect encrypted biometric data during matching processes, leading to potential leaks and security vulnerabilities, especially when repeated matching occurs or when malicious administrators are involved.
Innovation Solution
An encrypted text matching system that generates and uses auxiliary data to verify the Hamming distance between encrypted texts, performing one-way conversions on intermediate data to determine if the plaintexts match within a predetermined Hamming distance, ensuring secure matching without exposing original plaintext information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If encrypted text matching is performed using conventional methods, then matching functionality is achieved, but plaintext information may leak during the matching process
Solution Approach 1:
The patent introduces auxiliary data as an intermediary element that enables matching functionality without exposing plaintext information. The auxiliary data includes encrypted formants and encrypted Hamming distance values that act as mediators between the encrypted texts, allowing the matching process to proceed while maintaining security and preventing plaintext leakage.
Solution Approach 2:
The patent segments the encrypted text into multiple formants (first formant, second formant, etc.) and processes each segment independently through cryptographic operations. This segmentation allows the system to compute Hamming distances between segments without reconstructing or exposing the original plaintext, thereby maintaining security while enabling matching.
2Productivity
If repeated matching operations are performed, then authentication functionality is maintained, but security protection degrades over time
Solution Approach 1:
The patent implements periodic refreshment of auxiliary data including encrypted formants and encrypted Hamming distance values. By periodically updating these cryptographic elements while maintaining the same encrypted text, the system enables repeated authentication operations without degrading security, as each refresh creates new cryptographic material that prevents accumulation of exploitable information.
3Measurement precision
If auxiliary data is generated for Hamming distance verification, then matching accuracy is improved, but computational complexity increases
Solution Approach 1:
The patent performs preliminary cryptographic processing to generate auxiliary data (encrypted formants and encrypted Hamming distance values) during the enrollment phase. By pre-computing these elements before authentication, the system reduces the computational burden during repeated matching operations, as the auxiliary data can be reused and combined with new encrypted texts without requiring full re-computation each time.
Data Source
Figure 1
Figure 2(A)~2(B)
Figure 3
AI summary
The present invention makes it possible, in encrypted text matching, to avoid information leakage with regard to a plain source text and to ensure security. In the present invention, a system includes: a means for generating auxiliary data 1 and 2 for verifying that an encrypted text 1, which is input data that is encrypted and recorded in a storage device, and an encrypted text 2, which is input data to be matched that has been encrypted, match using the Hamming distance of the plain texts; and a means for unidirectionally converting at least one portion of the auxiliary data 2, unidirectionally converting intermediate data generated on the basis of the auxiliary data 1 and the difference between the encrypted text 1 and the encrypted text 2, and using the auxiliary data 2 and the results of the unidirectional conversion of the intermediate data to determine whether or not the Hamming distance of the plain texts corresponding to the difference between the encrypted text 1 and the encrypted text 2 is at or below a predetermined designated value.