ENCODING INFORMATION IN CHEMICAL CHAOS BY CONTROLLING SYMBOLIC DYNAMICS

In this paper we describe a technique of encoding and then decoding sy mbol sequences containing information into chaotic oscillations of the Belousov-Zhabotinsky reaction. The encoding technique is based on con trolling the chaotic oscillations by applying small parameter perturba tions and on learning the grammar of corresponding symbol dynamics pro duced by the free-running chaotic system. The use of small parameter p erturbations requires that we respect the grammar of the symbol dynami cs which represents the physical dynamical system. We present a method for learning the grammar of a symbol dynamics in terms of allowed tra nsitions between the bins defined by the symbol generating partition. The encoding technique can be easily utilized for targeting and stabil ization of any unstable periodic orbit.