Homomorphic Search in Encrypted Databases

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing homomorphic encryption methods for database querying are inefficient due to large key sizes and high computing costs, limiting their application to small datasets and lacking effective relevance ranking in information retrieval from large encrypted corpora.

Innovation Solution

A method for information retrieval in an encrypted corpus involves encrypting requests on a customer device, transmitting them to a server for homomorphic calculation, and decrypting the response on the device, with additional steps for calculating and updating term frequency and document frequency tables, and aggregating identifiers to present results in an orderly manner, utilizing parallel and distributed server calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If homomorphic encryption methods are used for database querying, then data confidentiality is improved, but computing cost and key size increase significantly

Engineering Contradiction:

Improvedata confidentialityVSAvoidcomputing cost

Core Design Contradiction:

ReliabilityVSUse of energy by stationary object

Solution Approach 1:

The patent segments the homomorphic encryption process into multiple components: encryption of search terms, encryption of document metadata (title, author, abstract), homomorphic matching operations, and result aggregation. This segmentation allows selective application of computationally intensive operations only where necessary, reducing overall computing cost while maintaining data confidentiality.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary encryption of document metadata (title, author, abstract) and organization of encrypted data structures before the actual search query is executed. This preliminary action prepares the encrypted corpus in advance, so that when a search query arrives, the homomorphic matching can proceed more efficiently without requiring extensive real-time computation.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If homomorphic encryption is applied to encrypted files on remote servers, then security against server compromise is improved, but search efficiency decreases

Engineering Contradiction:

Improvesecurity against server compromiseVSAvoidsearch efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The patent creates encrypted copies of document metadata (title, author, abstract) and stores them in structured formats on the remote server. These encrypted copies allow the server to perform homomorphic search operations without accessing or decrypting the actual document content, thereby maintaining security while improving search efficiency through optimized data structures and indexing.

Inventive Principle:

Principle #26Copying

3Use of energy by stationary object

If traditional encryption methods are used, then computing cost is reduced, but data confidentiality during querying is compromised

Engineering Contradiction:

Improvecomputing costVSAvoiddata confidentiality during querying

Core Design Contradiction:

Use of energy by stationary objectVSReliability

Solution Approach 1:

The patent changes the cryptographic parameters by using homomorphic encryption schemes that support arithmetic operations on encrypted data. This allows the system to perform search operations (matching, counting, ranking) directly on encrypted terms and metadata without decryption, maintaining data confidentiality while keeping computing costs manageable through optimized homomorphic operations.

Inventive Principle:

Principle #35Parameter changes

4Reliability

If homomorphic encryption is used for searching encrypted documents, then request confidentiality is improved, but the ability to rank results by relevance deteriorates

Engineering Contradiction:

Improverequest confidentialityVSAvoidrelevance ranking accuracy

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent introduces encrypted metadata (title, author, abstract) as an intermediary between the encrypted search query and the full document content. The homomorphic operations first match encrypted search terms against encrypted metadata, compute relevance scores based on metadata features, and only then retrieve and decrypt full documents for detailed examination. This intermediary approach enables relevance ranking while preserving request confidentiality.

Inventive Principle:

Principle #24Intermediary (Mediator)

Data Source

PatentUS11308233B2Method for information retrieval in an encrypted corpus stored on a server

Publication Date: 2022.04.19 WALLIX

AI summary

A method for information retrieval in an encrypted corpus stored on a server, from a digital request calculated on a customer device, containing a sequence of terms, includes the following steps: encryption of the request on a customer computer device and transmission of same to a database management server; and homomorphic calculation, on the server, of the encrypted response to the encrypted request recorded on the server. The method further comprises an additional requesting step performed on the customer device; and presentation of the result in an ordered form of the documents, in application of the processing of the previous step. The present disclosure also relates to a method for preparing a requestable base and to a method for information retrieval in an encrypted corpus.