Private Set Intersection via Homomorphic Encryption and Blinding
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for comparing private datasets between untrusted parties risk revealing sensitive information, as traditional comparison techniques lack adequate security measures to protect the underlying content of data items.
Innovation Solution
The implementation of homomorphic encryption and oblivious pseudo-random functions allows for the secure comparison of private data items by generating encrypted versions of identifiers, blinding them, and determining intersections without revealing the underlying content, while providing security guarantees to prevent unauthorized access or cheating.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional comparison techniques are used to determine intersections of private datasets, then the comparison process is simple and efficient, but the underlying content of data items may be revealed to the party performing comparisons
Solution Approach 1:
The patent introduces encrypted representations and oblivious pseudo-random functions as intermediaries between the two parties. Instead of directly comparing plaintext data items, each party encrypts their data items using the other party's public key, then applies oblivious PRF evaluations. This intermediary cryptographic layer allows comparison without revealing underlying content, resolving the contradiction between simple comparison and privacy protection.
Solution Approach 2:
The patent transforms data items from their original plaintext form into encrypted representations through cryptographic operations. By changing the parameter state of data from readable to encrypted form, and using homomorphic encryption properties, the system enables comparison operations on encrypted data without revealing the original values, thus protecting privacy while maintaining comparison functionality.
2Reliability
If encrypted comparisons are performed to protect data privacy, then data security is improved, but the computational overhead and processing time increase
Solution Approach 1:
The patent performs preliminary cryptographic setup operations including key pair generation and public key exchange before the actual comparison process. By preparing encrypted representations and establishing cryptographic parameters in advance, the system reduces the computational burden during the actual intersection determination, improving overall processing efficiency while maintaining security.
Solution Approach 2:
The patent replaces traditional mechanical comparison operations (direct plaintext comparison) with cryptographic operations (encrypted comparison using oblivious PRF). This substitution allows comparison to be performed on encrypted data without decryption, eliminating the need for secure key exchange during comparison and reducing overall computational overhead despite the initial encryption cost.
3Adaptability or versatility
If parties share data to benefit from collaboration, then mutual value is created, but the risk of unauthorized access to private information increases
Solution Approach 1:
The patent uses encrypted representations and oblivious pseudo-random functions as intermediaries that enable data sharing while preventing unauthorized access. Each party can contribute their data to the intersection computation without revealing the actual content to the other party. The cryptographic intermediaries ensure that only the intersection results (not the underlying private data) are disclosed, thus enabling collaboration while protecting against information leakage.
Data Source
AI summary
Systems, methods, devices, and other techniques for preserving privacy when comparing private datasets from first and second computing systems. The second computing system identifies a first set of identifiers corresponding to records in a private database of the second computing system. The second computing system receives blinded versions of a set of identifiers corresponding to records in a private database of the first computing system. The second computing system determines an intersection or characteristic thereof of the records in the private database of the first computing system and the records in the private database of the second computing system based on matches between the blinded versions of the first and second sets of identifiers.