Homophonic Coding Design for Communication Systems Employing the Encoding-Encryption Paradigm

Abstract: This paper addresses the design of a dedicated homophonic coding for a class of communication systems which, in order to provide both reliability and security, first encode the data before encrypting it, which is referred to as the encoding-encryption paradigm. The considered systems employ error-correction coding for reliability, a stream cipher for encryption, and homophonic coding to enhance the protection of the key used in the stream cipher, on which relies the security of all the system transmissions. This paper presents a security evaluation of such systems from a computational complexity point of view, which serves as a source for establishing dedicated homophonic code design criteria. The security evaluation shows that the computational complexity of recovering the secret key, given all the information an attacker could gather during passive attacks he can mount, is lower bounded by the complexity of the related LPN (Learning Parity in Noise) problem in both the average and worst case. This gives guidelines to construct a dedicated homophonic encoder which maximizes the complexity of the underlying LPN problem for a given encoding overhead. Finally, this paper proposes a generic homophonic coding strategy that fulfills the proposed design criteria and thus both enhances security while minimizing the induced overhead.