中文摘要：

由于缺少集中式服务器的全局管理,节点发现成为基于P2P的大规模分布式虚拟环境（DVE）设计的一个挑战.在满足全局连通性的基础上,通过满足AOI邻居封闭性、邻居封闭性和薄弱区周期关注性这三个不同强弱的拓扑要求来实现系统的局部感知性,提出了一种保证高拓扑一致性的节点发现分析框架.在此框架的基础上,设计了具有高拓扑一致性的完全分布式节点发现模型SimWorld.实验结果表明SimWorld模型的拓扑一致性接近100%.

英文摘要：

Due to the lack of global management of central server, node discovery becomes a challenge for P2P-based massively distributed virtual environment （DVE）. In order to solve such problem, an analytical framework for node discovery with high topology consistency is proposed. According to the framework, based on the respecting of global connectivity property, local awareness property can be maintained by fulfilling three topological requirements with different strengths, which are AOI neighborst encapsulation property, neighborsr encapsulation property and weak areas periodically checked property. And a fully distributed node discovery model with high topology consistency named SimWorld is designed founded on the framework. The results of the experiments show that the topology consistency of SimWorld model approximates to 100%.