作者: Craig Gentry , Sergey Gorbunov , Shai Halevi
DOI: 10.1007/978-3-662-46497-7_20
关键词: Graph (abstract data type) 、 Cryptography 、 Discrete mathematics 、 Computer science 、 Learning with errors 、 Directed graph 、 Key exchange 、 Multilinear map 、 Encryption 、 Multiplication
摘要: Graded multilinear encodings have found extensive applications in cryptography ranging from non-interactive key exchange protocols, to broadcast and attribute-based encryption, even software obfuscation. Despite seemingly unlimited applicability, essentially only two candidate constructions are known (GGH CLT). In this work, we describe a new graph-induced encoding scheme lattices. the arithmetic operations that allowed restricted through an explicitly defined directed graph (somewhat similar “asymmetric variant” of previous schemes). Our construction encodes Learning With Errors (LWE) samples short square matrices higher dimensions. Addition multiplication corresponds naturally addition LWE secrets. Security is not follow hardness (or any other “nice” assumption), at present it requires making assumptions.