Efficient homophonic coding
Boris Ya. Ryabko, Andrey N. Fionov
Abstract
Boris Ya. Ryabko, Andrey N. Fionov
Abstract
Homophonic coding, or homophonic substitution, is referred to as a technique that contributes to reliability of the secret key cipher systems. Its main goal is to convert the plaintext into a sequence of completely random (equiprobable and independent) code letters. In solving this problem three characteristics are to be considered: (i) redundancy, defined as the difference between the mean codeword length and the source entropy, (ii) an average number of random bits used in encoding, and (iii) complexity of the encoder and decoder, measured by memory size (in bits) and computation time (in bit operations). A class of homophonic codes is suggested for which both the redundancy and the average number of random bits can be made as small as required with nonexponential growth of memory size and roughly logarithmic growth of computation time.
OpenAlex reports 18 citations for this work. Citation counts describe recorded attention and do not establish research quality.
A contribution statement is not available in the OpenAlex record.
Method details are not available in the OpenAlex metadata.
Findings are not separately available in the OpenAlex metadata.
Limitations are not available in the OpenAlex metadata.
Application details are not available in the OpenAlex metadata.
Homophonic coding, or homophonic substitution, is referred to as a technique that contributes to reliability of the secret key cipher systems. Its main goal is to convert the plaintext into a sequence of completely random (equiprobable and independent) code letters. In solving this problem three characteristics are to be considered: (i) redundancy, defined as the difference between the mean codeword length and the source entropy, (ii) an average number of random bits used in encoding, and (iii) complexity of the encoder and decoder, measured by memory size (in bits) and computation time (in bit operations). A class of homophonic codes is suggested for which both the redundancy and the average number of random bits can be made as small as required with nonexponential growth of memory size and roughly logarithmic growth of computation time.
Key concepts: Code word, Encoder, Computer science, Algorithm, Decoding methods, Plaintext, Entropy (arrow of time), Redundancy (engineering)