计算机应用 ›› 2010, Vol. 30 ›› Issue (3): 733-735.

• 网络与通信 • 上一篇    下一篇

无线传感器网络拓扑识别算法

张志勇1,胡光岷2   

  1. 1. 电子科技大学宽带光纤传输与通信网技术教育部重点实验室
    2. 电子科技大学 宽带光纤传输与通信网技术教育部重点实验室
  • 收稿日期:2009-09-02 修回日期:2009-10-25 发布日期:2010-03-14 出版日期:2010-03-01
  • 通讯作者: 张志勇

Novel topology inference algorithm for wireless sensor network

Guang-min HU2   

  • Received:2009-09-02 Revised:2009-10-25 Online:2010-03-14 Published:2010-03-01

摘要: 获取无线传感器网络(WSN)的拓扑信息对于网络规划和管理具有重要意义。针对采用数据融合机制的无线传感器网络(WSN),证明了以某一节点的父节点的数据成功传输为条件时,该节点丢包的条件概率最小;以上述结论为基础,提出了一种新的WSN拓扑识别算法,仅以端到端的测量信息作为依据,可以在不增加网络负载、无需中间节点协作的情况下,获得准确的拓扑结构。NS-2仿真结果表明了该算法的有效性。

关键词: 无线传感网络, 网络层析成像, 数据融合, 拓扑识别

Abstract: Information of Wireless Sensor Network (WSN) topology is significant to network planning and management. For the WSN based on the data aggregation communication paradigm, the authors proved that the conditional probability of data loss of one node was minimum given that the data of the parent node were successfully transmitted to the sink. Based on this conclusion, a novel algorithm to infer WSN topology was proposed. The algorithm was capable of capturing accurate topology for WSN. It used end-to-end measurements and did not incur any additional burden on the network. NS-2 simulation results show that the proposed algorithm has high accuracy.

Key words: Wireless Sensor Network (WSN), network tomography, data aggregation, topology inference