Secure Multi-Party Computation Intermediate Value Recovery
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Secure multi-party computation systems face high computational overhead, making it difficult to perform basic elementary functions in near-real-time, which is unacceptably slow for many applications, especially in secure cloud storage and database operations.
Innovation Solution
Recovering intermediate information from shared shares and using it to perform joint public functions with the aid of a centralized entity, such as a dedicated server, which preprocesses and postprocesses shares to speed up computations and reduce data traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If secure multi-party computation is used to perform joint public functions, then security is maintained, but computational overhead is high and computation time is unacceptably long
Solution Approach 1:
The patent segments the computation process into distinct phases: local computation of intermediate values by individual parties, centralized aggregation of these intermediate values, and final result computation. This segmentation allows parallel processing while maintaining security constraints, thereby improving computation speed without compromising security.
Solution Approach 2:
The patent introduces an optional centralized entity that acts as an intermediary to aggregate intermediate computation results from multiple parties. This intermediary facilitates efficient coordination and data sharing among parties, reducing the overall computation time while preserving the security properties of multi-party computation.
2Loss of information
If basic elementary functions are computed using secure multi-party computation, then confidentiality is protected, but near-real-time performance cannot be achieved
Solution Approach 1:
The patent applies preliminary action by having each party pre-compute intermediate values locally before the centralized aggregation step. This preprocessing reduces the computational burden during the main computation phase, enabling faster execution of basic elementary functions while maintaining confidentiality through the secure multi-party computation framework.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
The invention relates to a method for obtaining a result of a joint public function for a plurality of parties in a secure multi-party computation environment, comprising the steps of a) Providing input data of the parties, b) Generating shares from the provided input data, c) Sharing the generated shares among the parties according to a secret sharing scheme, d) Performing the joint public function with the shared shares, e) Generating function shares from the performed public function and sharing the function shares among the parties according to a secret sharing scheme, and f) Obtaining the result from the shared generated shares, g) Recovering an intermediate information of the shared shares, and h) Using the recovered intermediate information for performing step d). The invention relates also to use of a method and a secure multi-party computation system.