Homophonic coding of cryptographic sources

We present a new algorithm for the homophonic coding of a message source for cryptographic purposes. The purpose of homophonic coding is to transform the output of a nonuniformly distributed message source into a uniformly distributed source sequence. This is achieved by randomly mapping each source symbol into one of a set of homophones. The selected homophones are then encoded by means of arithmetic coding. A new arithmetic coding algorithm is introduced, and simulation results show that, as expected, homophonic coding increases the entropy by less than 2 bits per symbol