Pattern recognition of one-dimensional cellular automata using Markov chains

A technique is presented for the identification of rule that generates a given complex pattern of linear one-dimensional cellular automata (LCA). The technique is based on the construction of a Markov transition matrix for the Markov chains that correspond to the evolution of the automaton. Such chain is generated by the evolution of a sequence of symbols representing the value of a string composed by small portion of the sites of the automaton. Excellent results are obtained for the identification of the rules that generate different complex patterns.