Efficient Instantiation of LWE-based Public-Key Encryption and Commitment Schemes

Abstract

The Learning with Errors (LWE) problem has been used as a underlying problem of a variety of cryptographic schemes. It makes possible constructing advanced solutions like fully homomorphic encryption, multi linear map as well as basic primitives like key-exchange, public-key encryption, signature. Recently, developments in quantum computing have triggered interest in constructing practical cryptographic schemes. In this thesis, we propose efficient post-quantum public-key encryption and commitment schemes based on a variant LWE, named as spLWE. We also suggest related zero-knowledge proofs and LWE-based threshold cryptosystems as an application of the proposed schemes. In order to achieve these results, it is essential investigating the hardness about the variant LWE problem, spLWE. We describe its theoretical, and concrete hardness from a careful analysis.