Evolutionary Games And IVCE Model Design Based On Complex Networks

Research of complex network theory is closely related to ecology, computer networks, and many other areas:in the area of ecology. Evolutionary dynamics on complex netowrks provide a theoretical framework for handling the central issue of evolutionary theory, and the research of network complexity can provide a useful theoretical basis for network resource optimization and large network topology design,Firstly, we analyzed previous research regarding evolutionary game on complex network, including the introduction of the concept of certain complex network statistic parameters, and introduced the construction method of some complex network models, and we introduced the basic models of evolutionary game theory.Secondly, we introduce a memory-based agent model and investigate the Prisoner's Dilemma game on a Heterogeneous Newman-Watts small-world network based on a Genetic Algorithm, focusing on heterogeneity's role in the emergence of cooperative behaviors. In contrast with previous results, we find that a different heterogeneity parameter domain range imposes an entirely different impact on the cooperation fraction. In the parameter range corresponding to networks with extremely high heterogeneity, the decrease in heterogeneity greatly promotes the proportion of cooperation strategy, while in the remaining parameter range, which relates to relatively homogeneous networks, the variation of heterogeneity barely affects the cooperation fraction. Also our study provides a detailed insight into the microscopic factors that contribute to the performance of cooperation frequency.Thirdly, we investigated the significance of cooperation evolution mechanims on the research of trust-based evolution game on large scale internet-based Virtual Computing Environment. Then based on analyzing the topology characteristics of the typical P2P network, we present a virtual computing topology constructing method, with load pressure based connection strategy, containing birth, growth, and maturity, besides supporting concurrent nodes joining and leaving. This method can generate network toplogy with more uniform degree distribution, and better load balancing capabilities. Simulation analysis showed that the degree distribution shows characteristics similar to normal distribution, and the network has a relatively small average path length and relatively larger clustering coefficient, indicating that the network has small-world properties.