无人值守传感器网络的高性能分布式存储算法  

Distributed storage algorithm with high performance for the UWSN

作　　者：肖宜龙[1,2] 蒋海波[3,2] 

机构地区：[1]中煤平朔集团有限公司,山西朔州036006 [2]中国科学院成都计算机应用研究所,四川成都610041 [3]中国科学院成都生物研究所,四川成都610041

出　　处：《西安电子科技大学学报》2014年第4期173-178,185,共7页Journal of Xidian University

基　　金：国家高技术研究发展计划资助项目(2008AAO1Z402);国家自然科学基金资助项目(61003034)

摘　　要：为解决无人值守传感器网络的数据存储可靠性问题,提出了一种具有低通信成本和低访问成本的分布式存储算法.算法采用步数为cn的并行定向随机游走机制,将网络中的k个源数据包按照一定的接收概率分散存储到网络中所有的n个节点,在每个节点形成一个存储数据包.理论分析和实验结果表明,基于该算法的存储过程完成之后,即使有部分传感器节点损坏,Sink节点只要随机收集到k+ε,ε大于等于11个存储数据包,就能成功地计算出原来的k个源数据包.与具有代表性的基于LT码的算法相比,文中算法将存储每个源数据包的通信次数从约3nlnn降至约3n;将读取源数据包的节点访问次数从大于k+100降至约k+11.To solve the data storage reliability problem of the unattended wireless sensor network consisting of n sensor nodes,this paper proposes a new distributed storage algorithm based on parallel directional random walk with length cn.The proposed algorithm disseminates each of k source data packets in the network according to a given probability and formes a stored data packet in every node.Simulation results show that,after the storage process based on the proposed algorithm is completed,even if some stored data packets are missing,the data collector node can successfully recover the k source data packets from any survival k＋ε,ε≥ 1 1 stored data packets.Compared with the LT codes based algorithm,the proposed algorithm reduces Sink node’s query cost from k＋100 to k＋11 and reduces the network’s communication cost from 3nlnn to 3n.

关 键 词：传感器网络 数据存储 分布式算法 随机游走 

分 类 号：TP211.9[自动化与计算机技术—检测技术与自动化装置] TP393[自动化与计算机技术—控制科学与工程]