作者: Shi Bai , Steven D. Galbraith
DOI: 10.1007/978-3-319-08344-5_21
关键词:
摘要: We consider the binary-LWE problem, which is learning with errors problem when entries of secret vector are chosen from { 0, 1} or − 1, 1 }. Our main result an improved lattice decoding algorithm for binary-LWE, by translating to inhomogeneous short integer solution (ISIS) and then re-scaling lattice. also discuss modulus switching as approach problem. conclusion that easier than general LWE. give experimental results, theoretical estimates parameters achieve certain security levels.