Secure Distributed Storage via Multiparty Secret Sharing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Distributed storage solutions for confidential data, particularly biometric data, face security challenges as they are often insecure, allowing unauthorized access by third parties, and existing solutions complicate access, making it time-consuming and costly.

Innovation Solution

A method for secure distributed storage and access using electronic circuits to obtain approximated short representations of confidential data, dividing them into shares and short parts, and transmitting these across multiple servers, ensuring only a threshold number of servers can reconstitute the data, while allowing rapid and secure identification through distance calculations.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If confidential data is stored in distributed form on multiple servers, then data availability and reliability are improved, but security is worsened as servers may access or reconstitute the data

Engineering Contradiction:
Improvedata availabilityVSAvoidunauthorized access
Core Design Contradiction:
ReliabilityVSObject-affected harmful factors

Solution Approach 1:

The confidential data is divided into N shares using secret sharing schemes, where each share is stored on a different server. No single server or small group of servers can reconstitute the original data, thus maintaining security while distributing storage across multiple locations for improved availability and reliability.

Inventive Principle:
Principle #1Segmentation

2Object-affected harmful factors

If cryptographic techniques are used to protect access to distributed confidential data, then security is improved, but access time increases significantly

Engineering Contradiction:
Improvedata protectionVSAvoidaccess time
Core Design Contradiction:
Object-affected harmful factorsVSLoss of time

Solution Approach 1:

The patent extracts and stores approximated short representations of the confidential data separately from the full data shares. During access operations, these compact representations enable rapid filtering and identification of relevant data before full decryption and retrieval, significantly reducing access time while maintaining security through the threshold-based share requirement.

Inventive Principle:
Principle #2Taking out (Extraction)

3Measurement precision

If full confidential data is downloaded for identification and search operations, then identification accuracy is improved, but bandwidth consumption and processing time increase

Engineering Contradiction:
Improveidentification accuracyVSAvoidbandwidth consumption
Core Design Contradiction:
Measurement precisionVSLoss of energy

Solution Approach 1:

Instead of downloading complete confidential data for identification operations, the system uses partial information in the form of approximated short representations stored on servers. These compact representations enable efficient filtering and candidate identification with minimal bandwidth consumption, while the full accuracy is restored only when needed through secure reconstitution from sufficient shares.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentEP2953291B1Secured distributed storage by multiparty calculation
Publication Date: 2019.11.20 IDEMIA IDENTITY & SECURITY FRANCE SAS
  • EP2953291B1 patent drawingFigure 1
  • EP2953291B1 patent drawingFigure 2
  • EP2953291B1 patent drawingFigure 3A

AI summary

The description relates in particular to a secure distributed storage method, a method for secure access to distributed storage, and devices, systems, computer programs and storage media for the implementation of such methods.