Secure Computation Apparatus Reducing Multiplication Operations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional secure computation techniques require a large number of additions/subtractions, multiplications, and memory accesses, which hampers their efficiency and arithmetic speed.
Innovation Solution
The technique involves secure computation apparatuses sharing subshares of values a and b across three secure computation apparatuses, using arithmetic units to calculate the result with reduced operations by sharing arbitrary values r+, and r-, thereby minimizing the number of additions, multiplications, and memory accesses.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional secure computation techniques are used, then security is maintained, but the number of additions/subtractions, multiplications, and memory accesses increases
Solution Approach 1:
The patent segments the computation process into distinct phases: sharing arbitrary values r+ and r- between apparatuses, distributing subshares a+、a-、b+、b- across different secure computation apparatuses, and performing localized calculations. This segmentation reduces the need for frequent inter-apparatus communication and memory accesses while maintaining security through distributed secret sharing.
Solution Approach 2:
The patent performs preliminary actions by pre-sharing arbitrary values r+ and r- between the secure computation apparatuses before the main computation. This preliminary setup allows the apparatuses to perform efficient local calculations without requiring repeated communication during the computation process, thereby improving arithmetic speed while maintaining security.
2Reliability
If conventional secure computation techniques are used, then security is maintained, but the number of memory accesses increases
Solution Approach 1:
The patent segments data storage and access patterns by distributing subshares across different apparatuses and organizing them into distinct groups (a+、a-、b+、b-). This segmentation allows each apparatus to access only the specific subshares it needs for its local calculations, reducing overall memory access frequency and time while maintaining security through distributed storage.
Solution Approach 2:
The patent performs preliminary action by pre-distributing and organizing subshares a+、a-、b+、b- across the secure computation apparatuses before the main computation begins. This preliminary distribution ensures that each apparatus has immediate access to the required data locally, minimizing the need for repeated memory accesses during computation and reducing time loss.
Data Source
AI summary
Pi and P+ have stored a+∈{a0, a1, a2} and b+∈{b0, b1, b2} therein, and Pi and P− have stored a−∈A− and b−∈B− therein. Here, P+−P(i+1)mod 3, P−=P(i−1)mod 3, and a and b are arbitrary values and satisfy a=a0+a1+a2 and b=b0+b1+b2, where A− is a complement of a+ in {a0, a1, a2} and B− is a complement of b+ in {b0, b1, b2}. Pi and P+ share r+, Pi and P− share r−, and Pi calculates c+=(a++a−)(b++b−)−a−b−+r+−r−. Pi sends c+ to P+.