Post-ModRaise FHE Key Switching With Aggregated MultSum
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Fully Homomorphic Encryption (FHE) schemes face significant computational and memory challenges due to inefficient key-switching operations, particularly after the ModRaise process, leading to memory bottlenecks, high computational demands, and scalability issues that hinder real-time commercial applications.
Innovation Solution
Optimize key-switching operations by pre-computing aggregated polynomial sums (Sum0 and Sum1) and simplifying the MultSum process, reducing polynomial multiplications from 2d to 2, and dynamically generating Sum1 from a seed value to minimize memory usage and computational overhead.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If traditional key-switching operations are performed after ModRaise in FHE, then ciphertext transformation is achieved, but computational cost and memory usage increase significantly
Solution Approach 1:
The patent applies preliminary action by pre-computing aggregated polynomial sums (Sum0 and Sum1) during the key generation phase rather than computing them during each key-switching operation. This pre-computation stores results in advance, eliminating redundant calculations during runtime and significantly reducing computational cost when performing multiple key-switching operations after ModRaise
Solution Approach 2:
The patent merges multiple polynomial multiplications into a single aggregated operation. By combining 2d polynomial multiplications into 2 aggregated polynomial sums (Sum0 and Sum1), the system reduces the number of separate computational operations, thereby lowering overall computational cost and improving key-switching efficiency
2Productivity
If traditional key-switching operations are performed after ModRaise in FHE, then ciphertext transformation is achieved, but memory usage increases due to large auxiliary data
Solution Approach 1:
The patent extracts and pre-computes only the essential aggregated polynomial sums (Sum0 and Sum1) that are necessary for key-switching operations. By extracting only the critical components needed for the operation and storing them in advance, the system minimizes auxiliary data size while maintaining the ability to perform efficient key-switching operations
Solution Approach 2:
By performing preliminary computation of aggregated polynomial sums during key generation, the system eliminates the need to store large amounts of intermediate data during runtime. This pre-computation approach reduces auxiliary data size by only storing the final aggregated results rather than all intermediate computational elements
3Measurement precision
If polynomial multiplications are performed individually in MultSum operation, then accuracy is maintained, but computational overhead increases
Solution Approach 1:
The patent merges 2d individual polynomial multiplications into 2 aggregated polynomial sum operations. By combining multiple multiplication operations into aggregated sums (Sum0 and Sum1), the system maintains computational accuracy while significantly reducing the number of operations required, thereby lowering computational overhead and time loss
Data Source
AI summary
A method and system of the device may include identifying, in an FHE program, key-switching operations occurring after a ModRaise operation. In addition, the device may include determining a first aggregated polynomial sum and a second aggregated polynomial sum; and configuring instructions for performing a MultSum operation based on the determined first aggregated polynomial sum and the second aggregated polynomial sum, the instructions are programmed to be executed during runtime, where the MultSum operation outputs two polynomials, each of which is an inner product between a single cyphertext polynomial and the first aggregated polynomial sum and a single cyphertext polynomial and the second aggregated polynomial sum, thereby reducing memory usage and computational overhead and enhancing key-switching efficiency.