Federated Ciphertext Query Indexing for Private Range and Ranking Search
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In joint data processing scenarios involving multiple data parties, existing methods struggle to preserve data privacy while efficiently querying data from a federated database, particularly in range and ranking queries, as they often expose sensitive information through database indexing.
Innovation Solution
A federated data query method and apparatus that uses a trusted third party with a cryptographic computing architecture to create a ciphertext database, disorder the data table, and generate indexes based on attribute value sorting, ensuring privacy by using coded row identifiers and secure multi-party computation to prevent data leakage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a third party constructs a ciphertext database for federated data processing, then data security is improved, but query efficiency deteriorates due to the need to repeatedly decrypt and process ciphertext data
Solution Approach 1:
The patent applies preliminary action by pre-processing the ciphertext data during database construction. The third party performs sorting and indexing operations on the ciphertext data in advance, creating an ordered ciphertext database with pre-computed indexes. This allows subsequent queries to leverage the pre-established order and indexes without requiring repeated full decryption and re-processing, thus improving query efficiency while maintaining data security through encrypted storage
Solution Approach 2:
The patent replaces traditional mechanical decryption-query cycles with a cryptographic ordering system. Instead of mechanically decrypting data for each query operation, the system uses cryptographic comparison and ordering mechanisms that operate directly on ciphertext. The third party establishes cryptographic order relationships between ciphertext values, enabling efficient range queries and rankings without plaintext decryption, substituting mechanical processing with cryptographic operations
2Speed
If traditional database indexing is used on ciphertext data, then query speed is improved, but data privacy is compromised through index leakage
Solution Approach 1:
The patent applies local quality by creating different types of indexes for different query scenarios. For equality queries, traditional equality indexes on ciphertext are used. For range queries, the system creates cryptographic range indexes that maintain order relationships without exposing plaintext values. For ranking queries, cryptographic ranking indexes are established. Each index type is optimized for its specific query pattern while maintaining appropriate privacy protection levels for that local context
Solution Approach 2:
The patent introduces cryptographic order relationships as an intermediary between the ciphertext data and the indexing mechanism. Instead of directly indexing plaintext values (which would leak privacy) or creating inefficient indexes on raw ciphertext, the system establishes cryptographic ordering relationships that mediate between the encrypted data and the index structure. This intermediary layer enables efficient querying while preventing direct exposure of plaintext information through the index
3Reliability
If data is disordered to prevent index leakage, then data privacy is improved, but query efficiency deteriorates due to loss of order information
Solution Approach 1:
The patent applies parameter changes by transforming the ordering parameter from plaintext values to cryptographic order relationships. Instead of ordering data based on plaintext attribute values (which would expose privacy if indexed), the system orders ciphertext data based on cryptographic comparison results. The order parameter is changed from direct plaintext representation to encrypted ordinal relationships, enabling privacy-preserving range queries and rankings while maintaining query efficiency through the preserved cryptographic order structure
Data Source
AI summary
Data query methods, apparatuses and computer-readable media are described. In an example process of processing, by a third party, a federated data table formed based on data of a plurality of data parties, if related data need to be obtained based on sorting of multiple attribute columns of the data table, when the federated data table is sorted based on attribute values of the attribute columns, a row identifier is introduced for an out-of-order data table obtained by disordering the federated data table, and an index is created based on the row identifier. The row identifier is determined by the third party and exists in a form of ciphertext in an index table. A row identifier of a candidate row can be restored to a plaintext.