Research On Evolution And Emergence Mechanism Of The Complex System Based On Cellular Automata

Complexity science,which is a multidisciplinary and frontier science,takes the complex system as a research object.The development of complex science will bring breakthroughs and innovations to the methodology.The study of the evolution and emergence mechanism of the complex system is an important part of the complex system.The basic model of the complex system is the cellular automata.The research on the complex evolution behavior and emergence mechanism of cellular automata is helpful to revealing the complex exploration and the study of inherent law of the complex system.In this paper,the evolution and emergence mechanism of the complex system based on cellular automata are studied,and the cellular automaton's evolution behavior is applied to image encryption.The main work and contribution of this paper are as follows:(1)To study the evolution of cellular automata by the evolutionary and the evolutionary mechanism's parameters.The evolutionary mechanism of elementary automata under different conditions is analyzed.It is found that the same cellular automata's evolution behavior had no substantial changes with different initial conditions.This paper presents a classification algorithm for two-dimensional cellular automata,which provides a great convenience for searching the complex rules of the two-dimensional cellular automata.Quantitative characterization of the evolutionary behavior of the different cellular automata is carried out by the existing power spectrum method.It is found that the self-organization ability is the root cause of the evolutionary complexity,and the stability of the power spectrum to different initial conditions is verified.The Langton parameter is not suitable for describing the evolution of all cellular automata,and there are multiple critical points,which are not conducive to searching for the cellular automata's different types of rules.On the basis of the Longton parameter,this paper proposes the equilibrium parameter,which is used to analyze the evolution law.In the experiment,by comparing with the Langton parameter.,It is found that the equilibrium parameter is more suitable and more suitable for searching for different types of rules.(2)By the emergence's characteristics to study the emergence mechanism.The existing method is to classify emergence phenomenon by the symmetrical breaking characteristics of regular elements.The existing method requires a lot of manpower to observe the characteristics of each rule by using the frequency map,and the need for each cell automata to write regular meta classification.In this paper,we propose the emergence phenomenon characteristics,which are the configuration density,the state's update frequency and the fractal dimension,and we propose a emergence phenomenon's classification method based on cost-sensitive support vector machine(CSVM).This method is used to verify the characteristics of emerging phenomena.The comparative experiment's results on the classification of the two cellular automata's emergence through the two method show that the method which we propose is more operational and has better expansibility,and the classification of cellular automata's emergence is more accurate.(3)The complex evolution behavior of cellular automata is applied to image encryption and decryption.An image encryption and decryption algorithm based on constructive reversible cellular automata and chaotic mapping is proposed.The algorithm does not need to transform the two-dimensional data into one dimension.By using the two-dimensional cellular automata classification algorithm in this paper,we can find the evolution rules with good randomness.The security of the encryption algorithm is analyzed.According to the analysis results,the encryption algorithm is safe and effective.