Secure multi-party computation methods, apparatuses, and systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional secure multi-party computation methods are hindered by the inability to combine homomorphic encryption and secret sharing protocols, leading to suboptimal performance and limited diversity in secure data processing.
Innovation Solution
Implement mutual conversion of encrypted data between homomorphic encryption and secret sharing, allowing these protocols to be used in combination to enhance secure multi-party computation performance and diversity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If homomorphic encryption and secret sharing are used as independent protocols, then each protocol can maintain its own security properties, but the performance and versatility of secure multi-party computation are limited
Solution Approach 1:
The patent combines homomorphic encryption and secret sharing protocols into a unified framework where encrypted data can be converted between the two representations. This merging allows the system to leverage the strengths of both protocols - the privacy preservation of secret sharing and the computational flexibility of homomorphic encryption - thereby improving versatility without requiring completely separate independent protocol implementations
Solution Approach 2:
The patent creates a universal conversion mechanism that allows encrypted data to function in both homomorphic encryption and secret sharing contexts. The conversion apparatus enables the same encrypted data to be adapted to different computational requirements, making the system multi-functional and able to handle diverse secure computation tasks with a single integrated approach
2Productivity
If only single protocol (homomorphic encryption or secret sharing) is used, then implementation is simpler, but computation efficiency and performance are suboptimal
Solution Approach 1:
The patent implements preliminary conversion of encrypted data from homomorphic encryption format to secret sharing format (or vice versa) before computational operations are performed. This preliminary action allows the system to optimize for the specific computational task at hand - using secret sharing for certain operations and homomorphic encryption for others - thereby improving overall computation efficiency while managing complexity through structured conversion processes
Solution Approach 2:
The patent introduces dynamic protocol selection and conversion capabilities that allow the system to adaptively switch between homomorphic encryption and secret sharing based on computational requirements. This dynamic approach enables optimization for different computational scenarios while maintaining a unified underlying architecture, improving productivity without requiring completely separate static protocol implementations
3Adaptability or versatility
If encrypted data cannot be converted between protocols, then protocol independence is maintained, but the diversity of secure data processing is limited
Solution Approach 1:
The patent introduces a conversion apparatus as an intermediary component that safely transforms encrypted data between homomorphic encryption and secret sharing formats. This intermediary maintains security properties by operating on encrypted data without requiring decryption, thereby preserving confidentiality while enabling protocol conversion. The intermediary acts as a controlled bridge that allows diversity in data processing while maintaining reliability through structured security-preserving transformations
Data Source
AI summary
Embodiments of this specification provide computer-implemented methods, apparatuses, computer-readable media, and systems for secure multi-party computation. In an example secure multi-party computation method, a first party encrypts a first plaintext segment of target data by using a homomorphic encryption algorithm based on a public key held by the first party in a first key pair to obtain a first ciphertext segment. A second plaintext segment of the target data is owned by a second party. The first party sends the first ciphertext segment to the second party. The second party performs a homomorphic addition operation in the homomorphic encryption algorithm on the first ciphertext segment and the second plaintext segment of the target data to obtain ciphertext data of the target data. The ciphertext data is decrypted based on a private key in the first key pair.