CAJ | 학술논문

在大规模节点密集的多跳传感器网络中，精确数据收集存在着“热区”问题：越靠近Sink节点的传感器节点，其承担的数据转发量就越多，能量消耗也越快，从而成为瓶颈节点，缩短整个网络的生命周期．最大生命周期数据收集树的构建已被证明是NP完全问题．已有算法大多是集中式算法，不适用于大规模节点密集的传感器网络．本文提出一种分布式精确数据收集算法EEDAT ，在大规模节点密集的传感器网络中，不仅能够保证每个节点到Sink的路径是最短路径（最少跳数），而且能有效延长网络生命周期．EEDAT分为两个基本步骤，首先随机生成一棵数据收集树，然后根据各个传感器节点的孩子数和剩余能量，对已生成的数据收集树进行调整，使得各个节点的负载尽量均衡，从而达到延长网络生命周期的目的．实验结果表明，与已有分布式算法LMST相比，EEDAT所构造的数据收集树能延长网络生命周期平均20％．
재대규모절점밀집적다도전감기망락중，정학수거수집존재착“열구”문제：월고근Sink절점적전감기절점，기승담적수거전발량취월다，능량소모야월쾌，종이성위병경절점，축단정개망락적생명주기．최대생명주기수거수집수적구건이피증명시NP완전문제．이유산법대다시집중식산법，불괄용우대규모절점밀집적전감기망락．본문제출일충분포식정학수거수집산법EEDAT ，재대규모절점밀집적전감기망락중，불부능구보증매개절점도Sink적로경시최단로경（최소도수），이차능유효연장망락생명주기．EEDAT분위량개기본보취，수선수궤생성일과수거수집수，연후근거각개전감기절점적해자수화잉여능량，대이생성적수거수집수진행조정，사득각개절점적부재진량균형，종이체도연장망락생명주기적목적．실험결과표명，여이유분포식산법LMST상비，EEDAT소구조적수거수집수능연장망락생명주기평균20％．
In this paper ,we propose a distributed tree-based algorithm for precise data gathering in wireless sensor networks called EEDAT ,which achieves longer network lifetime than existing distributed algorithms .The EEDAT algorithm has two steps .In the first step ,a shortest path tree is constructed in a distributed manner .In the second step ,EEDAT adjusts the load of nodes in the generated tree to balance energy consumption of different nodes ,which effectively extends the lifetime of the network .In the adjust-ment ,both the number of children of a node and its residual energy are considered .Simulation results show that EEDAT achieves longer lifetime than the LMST algorithm .On average ,compared with LMST ,EEDAT prolongs the lifetime by 20% .