Homomorphic Encryption for Anonymous Contact Graph Generation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing digital communication data collection methods fail to adequately protect user anonymity and privacy, as they often require decryption to analyze relationships, which compromises user identity and connection information.
Innovation Solution
A system utilizing homomorphic encryption allows users to create independent keys for encrypting contacts, with public keys shared among users and a central server computing encrypted equality computations to generate anonymous relationship graphs without revealing identities.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If traditional data collection methods are used to analyze user relationships, then relationship analysis can be performed, but user anonymity and privacy are compromised
Solution Approach 1:
Homomorphic encryption acts as an intermediary that allows the server to perform relationship analysis computations on encrypted contact data without being able to decrypt or view the actual user identities. The encryption scheme enables equality comparisons and graph computations while preserving user anonymity, thus resolving the contradiction between maintaining anonymity and enabling accurate relationship analysis.
Solution Approach 2:
The system changes the state of contact data from plaintext to encrypted form, transforming it into a different parameter state (ciphertext) that can still undergo mathematical operations. This parameter change allows the data to maintain its structural properties for relationship analysis while losing the ability to reveal actual identities, thus preserving anonymity without sacrificing analysis capability.
2Reliability
If user contact data is encrypted to protect privacy, then anonymity is preserved, but third-party analysis capability is reduced
Solution Approach 1:
Homomorphic encryption serves as an intermediary layer that enables third-party servers to perform analysis computations on encrypted data. The encryption scheme is designed to allow specific mathematical operations (equality comparisons, graph computations) to be performed on ciphertext without decryption, thus maintaining privacy protection while preserving third-party analysis capability.
Solution Approach 2:
The system replaces the traditional mechanical approach of decrypting data for analysis with a cryptographic approach where computations are performed directly on encrypted data. This substitution eliminates the need to compromise encryption for analysis purposes, as the cryptographic structure itself enables the required computations while maintaining security.
3Productivity
If centralized data collection is used for graph generation, then comprehensive relationship analysis is possible, but user control over personal information is lost
Solution Approach 1:
Homomorphic encryption acts as an intermediary that enables centralized graph generation while preserving user control. Users maintain control by generating their own key pairs and encrypting their contact data with their public keys. The centralized server can then process this encrypted data to generate comprehensive relationship graphs without gaining access to or control over the actual user information, thus maintaining both productivity and user control.
Data Source
AI summary
A public key generated by each user of a plurality of users is used to encrypt the contacts for that user. The results are sent to a server by each user. The key generated by each user is then distributed to every other user in the system, and each recipient encrypts their contacts with the keys. The result of these encryptions for all contacts for all recipients is then received by the server, and the server computes an encrypted computation of equality of two contacts and sends all computations back to the original user. The user can use the homomorphic property of the crypto protocol (e.g., a private key) to determine a set of users that are matched as contacts with the other users. The binary results are returned to the server, and the server computes a graph using the results.