Research on Homomorphic Encryption Methods for Financial Data Based on Parameter Optimization and Hybrid Architecture

Abstract

This study addresses the privacy protection requirements in financial data analysis by proposing a homomorphic encryption method that integrates CKKS parameter optimization with a hybrid FHE-PHE architecture. Through orthogonal experimental design, the optimal parameter combinations are selected, resulting in significantly improved encryption time compared to conventional encryption methods and greater stability in encryption time compared to FHE.