一种基于复杂网络的P2P流媒体拓扑构建算法

按照何种拓扑结构来构建覆盖网络是P2P流媒体网络中的热点问题,而已有的拓扑构建算法大多采用启发式算法,参数由经验值确定,缺乏理论分析。针对此问题,运用复杂网络中小世界效应的理论知识和簇结构的概念,提出了一种基于复杂网络的P2P流媒体拓扑构建算法,用所有的簇头节点构成一个小世界网络,利用小世界网络的特性来改善P2P流媒体网络的性能。对该算法和经典的CoolStreaming方法分别进行了仿真实验,比较了启动延迟、服务器负载、消息开销和抖动率等参数。实验结果表明新提出的算法相对于CoolStreaming方法具有显著的性能提高。复杂网络理论的运用是P2P流媒体系统研究的一个新方向,对今后的工作有重要的指导意义。

Topology formation algorithm for P2P streaming system based on complex network

Many interests have been put in choosing the right topological structure to establish overlay network in the study of the P2P streaming media technology. The existing topology formation algorithms are usually heuristic algorithms, in which the factors are determined from experience and lack of theoretical analysis. To solve this problem, this paper proposed a new algorithm for topological structure of P2P streaming media based on complex network, by applying the combined knowledge of cluster and complex network's small-world effect. This algorithm combined all the cluster head nodes to form a small-world network and improved the performance of P2P streaming media. Two simulation experiments have been conducted to compare the new algorithms and the existing CoolStreaming algorithm from four parameters: start-up latency, sever load, total message amount and jitter rate. The result shows that the new algorithm has a significant improvement in performance. The application of the complex network theory is a new trend of the P2P streaming media study. It's significant for the future work in the field.