Plaintext Index for Fully Homomorphic Encrypted Database Queries
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for fully homomorphic encrypted databases require excessive storage space and lead to delays in data processing speed, server transmission, and network load due to inefficient data encryption and storage practices.
Innovation Solution
A query information processing method that pre-builds a plaintext index for a fully homomorphic encrypted structured database, generates an encryption index through homomorphic encryption, and uses this index to efficiently process search queries, reducing the data size of search results and improving transmission efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing fully homomorphic encryption methods are used to store and encrypt data according to standard SIMD structure, then data can be operated on in encrypted form, but excessive storage space is required and the number of encrypted messages in search results increases
Solution Approach 1:
The patent segments the database operations into two distinct parts: (1) plaintext index structures that are built and maintained in unencrypted form for efficient storage and querying, and (2) encrypted data values that are stored separately. This segmentation allows the index to be compact and efficient while the encrypted data is protected, resolving the contradiction between encrypted operation capability and storage efficiency.
Solution Approach 2:
The patent introduces plaintext indexes as an intermediary structure that mediates between the encrypted data and the query processing system. The plaintext indexes enable efficient search and retrieval operations without requiring the encrypted data itself to be structured for indexing, thus reducing storage requirements while maintaining encrypted data operation capability.
2Quantity of substance
If the database size continuously increases with existing encryption methods, then more data can be stored, but data processing speed and server transmission efficiency are delayed
Solution Approach 1:
The patent performs preliminary action by pre-building and maintaining plaintext indexes before any query operations occur. These indexes are constructed in advance in an efficient plaintext format, allowing for rapid query processing as the database grows. This eliminates the need to process encrypted data during queries, thus maintaining high processing speed even as database capacity increases.
Solution Approach 2:
The patent changes the dimension of index storage by storing indexes in plaintext form in a separate dimensional space from the encrypted data. This dimensional separation allows the index structure to be optimized for speed and compactness independently of the encrypted data storage requirements, enabling the database to scale without proportionally increasing query processing time.
3Loss of information
If the number of encrypted messages in search results increases, then more complete search results can be provided, but network load and transmission delay increase
Solution Approach 1:
The patent extracts the indexing function from the encrypted data structure itself and places it in separate plaintext indexes. This extraction allows the system to provide complete search results by querying the compact plaintext indexes, then retrieving only the necessary encrypted data values. This dramatically reduces the number of encrypted messages that need to be transmitted compared to returning all encrypted data rows that match the query criteria.
Solution Approach 2:
The patent creates a plaintext copy of the index structure that mirrors the logical structure of the encrypted data without containing the actual encrypted values. This copy enables efficient query processing and result filtering, allowing the system to identify and retrieve only the specific encrypted data records that match the query, thereby reducing network transmission load while maintaining search result completeness.
Data Source
AI summary
According to the present disclosure, a query information processing method of a database performed on a computing device includes: pre-building a plaintext index corresponding to a fully homomorphic encrypted structured database; generating an encryption index by homomorphically encrypting the plaintext index; identifying query information requested from a client; indexing the encryption index in response to the query information; acquiring search result data through a homomorphic operation corresponding to a result value indexed by the encryption index; and providing the search result data to the client.