摘要
全同态加密算法可以实现密文状态下的数据运算,是解决数据隐私保护的主要技术之一.由于密文的噪音规模会随着运算增大,必须在噪音增大到一定规模之前进行噪音消减操作以保证密文可以正确解密,这一操作被称为自举操作.在目前的设计蓝图中,自举操作是实现全同态加密的核心技术,也是算法计算性能的瓶颈,阻碍了全同态加密算法的产业化应用.本文从自举的技术挑战入手,系统全面地分析了自举技术的现有解决思路和未来发展趋势.
Fully homomorphic encryption(FHE)can perform operations on the encrypted data,which is one of the main techniques to solve the problem of data privacy protection.Since the noise size of ciphertexts increases with the homomorphic operation,noise reduction operation must be performed before the noise size reaches a certain scale to ensure that the ciphertext can be decrypted correctly.This operation is called bootstrapping.In the current design blueprint,bootstrapping is the core technology to realize FHE,and it is also the bottleneck of computational performance,which hinders the industrial applications of FHE.Starting with the technical challenge of bootstrapping,this paper systematically and comprehensively analyzes the existing solutions and future development trends of bootstrapping.
作者
刘钦菊
路献辉
李杰
王鲲鹏
LIU Qin-Ju;LU Xian-Hui;LI Jie;WANG Kun-Peng(State Key Laboratory of Information Security,Institute of Information Engineering,Chinese Academy of Sciences,Beijing 100093,China;School of Cyber Security,University of Chinese Academy of Sciences,Beijing 100049,China)
出处
《密码学报》
CSCD
2021年第5期795-807,共13页
Journal of Cryptologic Research
基金
国家自然科学基金(61672030)。
关键词
隐私保护
全同态加密
自举
privacy protection
fully homomorphic encryption
bootstrapping
