摘要
针对目前基于LWE(Learn With Errors)构造全同态加密方案普遍需要高斯函数抽样、公钥尺寸过大等问题,提出利用高效的LWR(Learning With Rounding)替换传统的LWE,构造基于LWR(Learning With Rounding)的身份基全同态加密方案,对方案的安全性进行严格证明。LWR问题是LWE问题的变体,消除了LWE问题利用高斯函数抽样生成噪声的过程,具有更高的计算效率以及更小公钥和密文尺寸。
In order to solve the problem that the LWE based fully homomorphic encryption scheme generally needs samples of Gaussian function and oversized public key,a new LWR (Learning With Rounding) algorithm is proposed to replace the tradi tional LWE problem and construct a LWR With Rounding, and the safety of the program is strictly proved. The LWR problem is a variant of the LWE (Learn With Errors) problem,which eliminates the process of generating noise of LWE in using Gaussian function. The sampling has higher computational efficiency and public key and ciphertext is of smaller size.
作者
张兴兰
卢玉顺
ZHANG Xing-lan;LU Yu-shun(Department of Information,Beijing University of Technology,Beijing 100124,China)
出处
《软件导刊》
2018年第10期200-203,208,共5页
Software Guide
关键词
LWR
LWE
基于身份
全同态加密
高斯函数
LWR
LWE
identity based encryption
fully homomorphic encryption
Gaussian noise sampling