Homomorphic Ciphertext Sorting With Reduced Bootstrapping Overhead
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fully homomorphic encryption methods, such as TFHE, require excessive computational resources and time for operations on encrypted databases, making them impractical for real-world applications.
Innovation Solution
An encryption processing device and method that utilizes multi-value logical operations and reduced Gate Bootstrapping processes to efficiently rearrange and sort encrypted data within a database without decryption, using TLWE and TRLWE encryption techniques to minimize computational overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If fully homomorphic encryption (TFHE) is used to perform operations on encrypted data, then data security is improved, but computational complexity and processing time increase significantly
Solution Approach 1:
The patent segments the computational process by separating Bootstrapping operations (error reduction) from logical operations. The Bootstrapping process is performed periodically to reduce accumulated errors, while logical operations are performed in between. This segmentation allows the system to manage computational complexity by breaking down the overall processing into manageable stages rather than performing full Bootstrapping after every operation.
Solution Approach 2:
The patent performs Bootstrapping operations in advance before the error accumulation becomes critical. By proactively reducing errors at scheduled intervals rather than reactively when errors become unmanageable, the system maintains data integrity while avoiding the excessive computational burden of emergency error correction. This preliminary action approach optimizes the balance between security and computational efficiency.
2Reliability
If Bootstrapping is performed frequently to reduce error accumulation, then decryption reliability is improved, but processing time increases
Solution Approach 1:
The patent implements periodic Bootstrapping operations at predetermined intervals rather than continuously or after every logical operation. The system performs Bootstrapping on selected ciphertexts based on error threshold criteria, allowing error accumulation within acceptable bounds during intermediate steps. This periodic approach ensures decryption reliability is maintained while minimizing unnecessary computational overhead and processing time delays.
3Productivity
If multi-value logical operations are used to reduce calculation steps, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent designs the logical operation unit to handle multiple operation types (AND, OR, NOT, and multi-value operations) within a single unified structure. This universal operation unit can perform various logical functions by configuring the same hardware resources differently, rather than requiring separate dedicated circuits for each operation type. This multi-functionality approach increases processing speed and productivity while controlling device complexity through resource sharing.
Data Source
AI summary
An operation by a full adder that achieves fully homomorphic encryption is speeded up. Ciphertexts are ciphertexts of fully homomorphic encryption that allow a logical operation to be performed without decryption. A sorting process is performed which rearranges the ciphertexts arranged in a predetermined order in accordance with a predetermined method. In the sorting process, a homomorphic operation based on a first ciphertext and a second ciphertext is performed. From a result of this homomorphic operation and a ciphertext obtained based on a predetermined polynomial, new first and second ciphertexts are calculated. When the first and second ciphertexts and the new first and second ciphertexts satisfy a predetermined condition, the first ciphertext and the second ciphertext are swapped.