Higher Rates and Information-Theoretic Analysis for the RLWE Channel
share
The Learning with Errors (LWE) problem is an 𝒩𝒫-hard problem that lies the foundation of several cryptographic algorithms. Generalizing this principle, several cryptosystems based on the closely related Ring Learning with Errors (RLWE) problem have been proposed within the NIST PQC standardization process, e.g., the systems LAC and NewHope. The combination of encryption and decryption for these kinds of algorithms can be interpreted as data transmission over noisy channels. To the best of our knowledge this paper is the first work that analyzes the capacity of this channel. In particular, we present lower bounds on the capacity, which show that the transmission rate can be significantly increased compared to standard proposals in the literature. Furthermore, under the assumption of stochastically independent coefficient failures we give achievability bounds as well as concrete code constructions achieving a decryption failure rate (DFR) less than 2^-112 for LAC and less than 2^-216 for NewHope for their highest proposed security levels. Our results show that the data rate of the encryption scheme is significantly increased for NewHope, namely by factor of 7 and for LAC by a factor of 2. This is partly based on choosing larger alphabet sizes. Furthermore we show how to significantly decrease the decryption failure rate compared to the original proposals while achieving the same bit rate. We also show that the measures we take to achieve these results have no negative impact on the security level. Thus, by means of our constructions, we can either increase the total bit rate while guaranteeing the same DFR or for the same bit rate, we can significantly reduce the DFR (e.g., for NewHope from 2^-216 to 2^-12764).
READ FULL TEXT