XOR-Based Data Division for Large-Scale Secret Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data division methods, such as those using threshold secret sharing schemes with polynomial and residue calculations, become impractical for large data sets like Megabytes or Gigabytes due to high computational requirements and memory needs, and lack flexibility in processing unit selection.
Innovation Solution
A data division method that divides original data into desired number of parts using a prescribed processing unit bit length through exclusive OR calculations with randomly generated partial data, allowing recovery from a subset of divided data without requiring high computational power or proportional memory capacity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional polynomial and residue calculation methods are used for data division, then data secrecy and recovery capability are improved, but computational processing power and memory capacity requirements increase linearly with data size
Solution Approach 1:
The patent replaces complex polynomial calculations and residue calculations with simple bitwise XOR operations. Instead of using mathematical fields and polynomial arithmetic, the invention uses binary bit manipulation where each bit of the divided data is generated by XORing corresponding bits of the original data with random bits. This substitution dramatically reduces computational complexity from O(n^2) or higher to O(n), making large-scale data division practical.
Solution Approach 2:
The patent changes the fundamental parameter of calculation complexity by switching from algebraic operations (polynomial multiplication and modular arithmetic) to logical operations (bitwise XOR). This parameter change transforms the computational burden from heavy mathematical processing to lightweight bit manipulation, enabling efficient processing of large datasets while maintaining the threshold secret sharing property.
2Reliability
If conventional polynomial and residue calculation methods are used for data division, then data secrecy and recovery capability are improved, but memory capacity requirements increase linearly with the number of divisions
Solution Approach 1:
The patent replaces complex polynomial calculations and residue calculations with simple bitwise XOR operations. Instead of using mathematical fields and polynomial arithmetic, the invention uses binary bit manipulation where each bit of the divided data is generated by XORing corresponding bits of the original data with random bits. This substitution dramatically reduces computational complexity from O(n^2) or higher to O(n), making large-scale data division practical.
Solution Approach 2:
The patent changes the fundamental parameter of calculation complexity by switching from algebraic operations (polynomial multiplication and modular arithmetic) to logical operations (bitwise XOR). This parameter change transforms the computational burden from heavy mathematical processing to lightweight bit manipulation, enabling efficient processing of large datasets while maintaining the threshold secret sharing property.
3Device complexity
If fixed polynomial degree is used for threshold secret sharing, then mathematical structure is simplified, but flexibility in processing unit selection is lost
Solution Approach 1:
The patent introduces dynamic adaptability by allowing the processing unit size to be flexibly selected based on requirements. The system can process data in units of bits, bytes, or larger blocks, and can adjust the threshold parameter k independently from the total number of divisions n. This dynamic configuration enables the same XOR-based framework to adapt to different security requirements and data sizes without restructuring the underlying mathematical framework.
Solution Approach 2:
The patent segments the data processing into independent bitwise operations that can be applied at different granularities. By dividing the original data into multiple sets of divided data where each set contains the same number of bits, the system allows flexible selection of processing unit size. Each bit position can be processed independently through XOR operations, enabling parallel processing and flexible adaptation to different computational resources and security requirements.
Data Source
AI summary
A plurality of original partial data are generated by dividing the original data by the prescribed processing unit bit length, a plurality of random number partial data each having a length less than or equal to the prescribed processing unit bit length are generated in correspondence to the plurality of original partial data, and a plurality of divided partial data that constitute each divided data are generated by using exclusive OR calculation of the original partial data and the random number partial data, each divided partial data having a length equal to the prescribed processing unit bit length. Then, the divided data in the desired number of division are generated from the plurality of divided partial data, such that the original data cannot be ascertained from any one divided data alone but the original data can be recovered from a prescribed number of the divided data among generated divided data.