Cellular Automata: Optimal Analysis, Coding, and Testing for Encryption

Abstract

Cellular automata are a set of discrete structures generated and manipulated by predetermined rules, in which each state (or evolution) is influenced by the previous. Utilizing the simplicity of this fundamental structure, a number of configurations have been organized and derived from elementary (single dimensional) cellular automata. By harvesting the evolution of these structures as output, they lend greatly to random number generation and by extension, encryption. Analyzing, testing, and programming these methods has led to observations on optimal approaches to each. Utilizing the Diehard testing suite and the National Institute of Standards and Technology (NIST) Statistical Testing Suite (STS), configurations can be judged against each other as well as external systems. Optimal methods for generating configurations, visual observation and data analysis are compiled in a workbook program. A complete analysis for the state diagrams of k [1, 27] in the 3-bit rule space is included and a Cellular Automata Standard of Encryption (CASE) is suggested for real world use.