Encrypted Data Search Indexing via Keyword Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional secure search systems using probabilistic encryption face significant performance issues, as they are unable to perform fast searching of encrypted data when the keyword has not been searched for previously, due to the need to compare all stored encrypted tags with the search query, leading to slow search times.

Innovation Solution

A data processing apparatus that specifies a keyword for storage, generates a bit sequence, and performs concealing processing to disclose only certain bit positions to the data storage apparatus, allowing for the derivation of a storage index value to be attached to tag data, enabling fast searching without revealing sensitive information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If probabilistic encryption is used for secure search, then security is improved because frequency analysis cannot be performed, but search speed deteriorates because all encrypted tags must be compared with the search query

Engineering Contradiction:
ImprovesecurityVSAvoidsearch speed
Core Design Contradiction:
ReliabilityVSSpeed

Solution Approach 1:

The patent segments the encrypted tag into multiple parts, each associated with different keywords. When a search is performed, only the relevant segment needs to be compared with the search query, rather than comparing the entire tag or all tags. This segmentation enables faster search while maintaining the security benefits of probabilistic encryption.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent performs preliminary processing to create an index structure that organizes encrypted tags by keyword segments before search operations. This preliminary organization allows the system to quickly locate and compare only the relevant encrypted tags during search, rather than performing exhaustive comparisons across all stored tags.

Inventive Principle:
Principle #10Preliminary action

2Speed

If deterministic encryption is used for secure search, then search speed is improved because high-speed database methods can be used, but security deteriorates because frequency analysis can infer data content

Engineering Contradiction:
Improvesearch speedVSAvoidsecurity
Core Design Contradiction:
SpeedVSReliability

Solution Approach 1:

The patent applies local quality by using deterministic encryption only for specific segments of the tag that are used for indexing and search comparison, while other parts of the data remain encrypted with probabilistic encryption. This allows fast search operations on the deterministic segments while maintaining security through probabilistic encryption of the remaining data.

Inventive Principle:
Principle #3Local quality

3Reliability

If all encrypted tags are compared with search query, then security is maintained by using probabilistic encryption, but search time increases significantly for first-time queries

Engineering Contradiction:
ImprovesecurityVSAvoidsearch time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent extracts and separates the keyword-associated segments from the full encrypted tags, storing them in an optimized index structure. During search operations, only these extracted keyword segments are compared with the search query, dramatically reducing the comparison workload while maintaining the security properties of the full probabilistic encryption scheme.

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentEP2665052B1Data processing device and data archiving device
Publication Date: 2018.08.15 MITSUBISHI ELECTRIC CORP
  • EP2665052B1 patent drawingFigure 1
  • EP2665052B1 patent drawingFigure 2
  • EP2665052B1 patent drawingFigure 3

AI summary

An access terminal apparatus 301 provides a group determination key being a decryption key to a data center apparatus 401 previously, and then transmits grouping information generated from a keyword of storage target data and having been encrypted to the data center apparatus 401, with encrypted data of the storage target data and tag data of the encrypted data of the storage target data. The data center apparatus 401 stores the encrypted data relating it to the tag data, decrypts a part of the grouping information with the group determination key, derives an index value from a bit value obtained by the decryption, and stores the derived index value and the tag data relating them with each other.