摘要：

传感器节点容易发生软件或机械故障，从而导致节点数据部分或全部丢失。为保证无线传感器网络数据收集的可靠性，提出一种改进的数据在线恢复方案。采用分布式存储机制对节点数据进行冗余处理，依据节点发出的恢复请求，给出改进的多项式时间数据恢复算法，支持请求接纳控制，并对数据恢复文件块大小进行限制，可在非循环网络上取得恒定的近似率边界。理论分析与仿真实验结果表明，该方案可准确恢复网络数据，对各数据恢复请求均可实现数据恢复成本最小化。

关键词: 传感器节点, 节点故障, 数据恢复, 分布式存储, 近似率, 成本

Abstract:

Since the sensor nodes prone to the software or mechanical failure, this may cause the permanent data loss of some or all of the data. In order to ensure the reliability of data collection in Wireless Sensor Network(WSN), this paper proposes an improved data online recovery scheme. The data of nodes are redundant processed by using the distributed storage mechanism, and based on the recovery request sent by the sensors, a constant approximation competitive ratio polynomial time data recovery algorithm is proposed, which can have the low complexity competitive ratio when admission control is allowed, and the size of recovery pieces is constrained, and achieves the constant approximation ratio bound in acyclic networks. Analysis and simulation experimental results show that the proposed algorithm can recover network data exactly and attain minimum recovery cost for any recovery request.

Key words: sensor node, node failure, data recovery, distributed storage, approximation ratio, cost

中图分类号: 

• TP393