Encrypted Data Retrieval System with Dynamic Re-Encryption
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing technologies for retrieving encrypted data do not provide sufficient confidentiality, as the retrieval result is not concealed, allowing the keyword used for retrieval to be deciphered from the frequency of the retrieval result.
Innovation Solution
A retrieval system that uses a provider machine to perform encrypted queries on encrypted data, with a first management unit that decrypts and re-encrypts the retrieval keyword and ID in a Trusted Execution Environment (TEE) to prevent administrator access, and a second management unit that updates index information with the re-encrypted values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If data is stored and retrieved in encrypted form using conventional encryption methods, then data confidentiality is improved, but the retrieval result becomes decipherable through frequency analysis
Solution Approach 1:
The patent applies dynamics by making the encryption process adaptive and changing over time. Each retrieval operation triggers a re-encryption of the retrieval keyword and updates the index, so the encryption parameters are not static but dynamically adjusted based on usage patterns. This prevents frequency analysis attacks that rely on consistent encryption patterns.
Solution Approach 2:
The patent changes encryption parameters dynamically during retrieval operations. The retrieval keyword is re-encrypted with updated parameters, and the index is refreshed accordingly. This parameter transformation ensures that even repeated queries for the same data produce different encrypted forms, eliminating frequency analysis vulnerabilities.
2Productivity
If the retrieval keyword and ID are stored in plaintext or consistently encrypted form, then retrieval efficiency is improved, but the frequency of retrieval results can be analyzed to decipher keywords
Solution Approach 1:
The patent performs preliminary re-encryption of the retrieval keyword and updates the index before the actual retrieval operation completes. This preliminary action ensures that by the time retrieval results are returned, the encryption parameters have already been updated, preventing any window where consistent encryption patterns could be exploited for frequency analysis.
Solution Approach 2:
The patent introduces an intermediary re-encryption step between the initial encrypted query and the final retrieval result. This intermediary process transforms the encryption parameters, acting as a mediator that breaks the direct link between query patterns and result frequencies, thereby protecting keyword confidentiality while maintaining retrieval functionality.
3Ease of operation
If the administrator can access the storage area for management operations, then system management is simplified, but the security of encrypted data and retrieval keywords is compromised
Solution Approach 1:
The patent segments the storage system into multiple distinct storage areas with different access permissions. The first storage area contains sensitive encrypted data and retrieval keywords with restricted access, while the second storage area contains the index with controlled access. This segmentation allows the administrator to manage the index without accessing sensitive data, balancing management ease with security.
Solution Approach 2:
The patent applies local quality by assigning different security characteristics to different parts of the storage system. The first storage area has high security restrictions preventing administrator access, while the second storage area allows controlled access for index management. This localized security approach enables efficient administration of non-sensitive components while protecting sensitive data from exposure.
Data Source
Figure 1
Figure 2
Figure 3~5
AI summary
A retrieval system includes a first management unit which stores, in a first storage area that cannot be browsed by an administrator of a provider machine, management information in which an encrypted retrieval keyword used for a retrieval performed by the provider machine, and an encrypted ID, which is associated with the encrypted retrieval keyword, are associated, and, of the management information, decrypts, and then once again encrypts, the encrypted retrieval keyword corresponding to the encrypted query from a browsing machine and the encrypted ID corresponding to the encrypted retrieval keyword, and a second management unit which stores, in a second storage area, index information in which the encrypted retrieval keyword used for the retrieval performed by the provider machine, and one or more encrypted IDs associated with the encrypted retrieval keyword, are associated, and, of the index information, updates the encrypted retrieval keyword and the encrypted ID corresponding to the encrypted query from the browsing machine to the encrypted retrieval keyword and the encrypted ID which were decrypted and then encrypted by the first management unit.