针对P2P点播系统随机邻居节点选择算法大幅度降低网络带宽利用率以及网络吞吐量的问题,提出一种基于位置感知的邻居节点算法(LANS)构建邻居网络.依据位置感知技术计算出各个邻居节点与请求节点的网络位置,对相近的节点进行区域划分;将节点的在线时长、带宽、延时等参数综合成为节点的服务能力值,优先选择距离近、上传带宽大、在线稳定性好的邻居节点进行数据交互.利用PeerSim对该节点选择算法进行了仿真实验.实验结果表明,该算法能显著提高P2P点播系统的吞吐量,减少用户的下载时间,从而保证了P2P流媒体点播服务质量.%In order to reduce the utilization ratio of network bandwidth and network throughput,a neighbor node network al-gorithm based on location sensing is proposed.Firstly,the location of each neighbor node and the requesting node is calcu-lated according to the location sensing technology,and the neighboring nodes are divided into regions.Secondly,the param-eters such as the duration,bandwidth and delay of the node are integrated into the service capability of the node.We give preference to neighbor nodes for data interaction,which have the advantages of short distance,big upload bandwidth and good online stability.Finally,the simulation experiment of the node selection algorithm is carried out by using PeerSim. The experimental results show that the algorithm can significantly improve the throughput of P2P on demand system and re-duce the download time of users.The quality of P2P streaming service is improved.
展开▼
