MP2P Traffic Intelligent Management Model Based on BP neural network

摘要：

A Mobile Peer-to-Peer (MP2P) traffic intelligent management model (BP Neural Network and Intelligent Management, BP-IM) based on neural network is proposed. First, the whole MP2P network is modeled by the semi-distributed structure. Then the BP neural network is introduced to measure and control the MP2P flow effectively. In addition, the corresponding P2P traffic priority table is established and the assignment of the buffer flow is dynamically adjusted. The BP-IM model has the features of flexible configuration, efficient detection, and easy to be extended, reducing the complexity of the algorithm. Computer simulations based on OMNeT++ show that: compared to the traditional model, higher-level traffic processing delay is small by the proposed BP-IM model, which saves the limited MP2P bandwidth resources and avoids network congestion. MP2P traffic processing delay decreases with increase of the traffic priority, which reflects the intelligent traffic management concept for BP-IM model.

关键词： peer-to-peer traffic model intelligent management BP neural network

作者: Haitao Qu Meina Song Rihua Wang Binjie Zhu Wu Qu Junde Song

作者单位: ICT&SSME Center, Computer School, Beijing University of Posts and Telecommunications Beijing, China

会议类型: 国际会议

会议名称: The 2010 International Conference on Computer Application and System Modeling(2010计算机应用与系统建模国际会议 ICCASM 2010)

会议地点: 太原

会议语种:英文

页码: 360-364

在线出版日期: 2010-10-22（万方平台首次上网日期，不代表论文的发表时间）