Multi-Party Computation for Encrypted Network Value Aggregation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for monitoring user interactions with digital components lack security and privacy, as they are vulnerable to fraud and do not effectively protect user identity.
Innovation Solution
The implementation of a multi-party computation process using multiple aggregation servers to decrypt and attribute impression and conversion data, ensuring user privacy through double encryption and preventing fraudulent activity by enabling each server to check for signs of fraud.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If impression data and conversion data are stored and processed in plaintext by a single aggregation server, then the system is simple to implement and process, but user privacy is compromised and fraudulent activity can occur
Solution Approach 1:
The system divides the aggregation server into multiple separate servers (first aggregation server and second aggregation server). Each server processes only encrypted data and cannot access plaintext user information. The join keys are split between servers, requiring collaboration to match impressions with conversions, thereby maintaining privacy while enabling the full functionality.
Solution Approach 2:
The patent introduces encrypted join keys as an intermediary mechanism that allows the aggregation servers to link impression data with conversion data without ever decrypting the user-identifying information. The join keys act as a mediator that enables matching while preserving privacy, as only the encrypted forms are transmitted and processed between servers.
2Reliability
If data is encrypted using a single public key, then encryption and decryption are simple, but a single compromised server can access all user data
Solution Approach 1:
The patent combines multiple encryption layers by encrypting impression data with the first aggregation server's public key and conversion data with the second aggregation server's public key. This merging of encryption mechanisms ensures that no single server can access all data, as each server only holds one encryption key pair and must collaborate to process the full dataset.
3Measurement precision
If the system processes and matches impression and conversion data, then conversion tracking is accurate, but user privacy is exposed through data matching
Solution Approach 1:
The system creates encrypted copies of join keys that can be matched between impression and conversion data without revealing the actual user identifiers. The encrypted join keys serve as placeholders that enable accurate matching while the real user information remains encrypted and inaccessible to the servers, thus maintaining both accuracy and privacy.
Data Source
AI summary
Methods, systems, and apparatus, including a method for determining network measurements. In some respects, a method includes receiving, by a first aggregation server and from each of multiple client devices, encrypted impression data. A second aggregation server received from each of at least a portion of the multiple client devices, conversion data that includes, for each conversion recorded by the client device, encrypted conversion value data. The first aggregation server and the second aggregation server perform a multi-party computation process to decrypt the encrypted impression data and the encrypted conversion data.