A new type of keystream generator based on chaotic maps: illustration on a Henon generalized map

This paper presents a new way of using chaotic maps for the generation of pseudorandom sequences of good statistical properties, aiming to use them in cryptographical applications. The proposed pseudorandom sequence generator is a practical zero-memory information source, with a discrete set of possible values of varying size, obtained through successive random variable transformations applied to the chaotic map. We exemplify on a particular case of the generalized Henon map. The paper also presents an example of using a pseudorandom chaotic sequence in an enciphering chain, applied to images. The chaotic sequences are studied through conventional statistical methods and also using the NIST test suite. They satisfy the respective randomness requirements.