Search for a command to run...
Abstract Currently, the security problem of cloud data is becoming increasingly prominent. However, the existing information security protection methods pose a problem: they affect data query efficiency and cannot achieve a balance between encryption performance and encryption efficiency. To address this issue, the research proposed a Chinese Remainder Theorem-based additive homomorphism algorithm for secure query of ciphertext databases. The research first introduced the Chinese Remainder Theorem to optimize the key generation process of homomorphic encryption. This algorithm incorporated the Great Method of Seeking One Solution and transformed the mixed numbers into integer form, enabling the homomorphic encryption scheme to be applied to the fractional domain. The Bootstrapping technique was utilized to refresh the ciphertext and reduce noise accumulation. Based on the homomorphic encryption and the symmetric encryption, access control mechanism were combined to achieve multi-level encryption. Experimental results showed that the average encryption and decryption efficiency of the proposed method was 335.36kb/s and 321.28kb/s respectively, and the average query time was only 604.45ms, with an average query accuracy of 0.92. In practical application scenarios, the query delay of Method 1 was only 80.15ms, the ciphertext storage expansion rate was 8.47 times, the key switching time was 50.36ms, and the concurrent throughput was 736.14QPS. Compared with existing homomorphic encryption methods, the research achieved a better balance among security, efficiency and practicality. It extended homomorphic encryption to the fractional domain and real number domain, supporting a wider range of mathematical operation types, while maintaining the characteristic of ciphertext being directly computable.