An Observation on the Security of McEliece’s Public-Key Cryptosystem

Abstract

The best known cryptanalytic attack on McEliece’s public-key cryptosystem based on algebraic coding theory is to repeatedly select k bits at random from an n-bit ciphertext vector, which is corrupted by at most t errors, in hope that none of the selected k bits are in error until the cryptanalyst recovers the correct message. The method of determining whether the recovered message is the correct one has not been throughly investigated. In this paper, we suggest a systematic method of checking, and describe a generalized version of the cryptanalytic attack which reduces the work factor significantly (factor of 2¹¹ for the commonly used example of n=1024 Goppa code case). Some more improvements are also given. We also note that these cryptanalytic algorithms can be viewed as generalized probabilistic decoding algorithms for any linear error correcting codes.

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.