【作者】 李伟；

【导师】 任秀丽；

【作者基本信息】 辽宁大学 ， 计算机应用技术， 2020， 硕士

【摘要】 无线传感器网络利用部署在监控环境中的节点监测不同区域位置用户需求的目标信息,最终向用户反馈相关数据信息,目前被广泛应用于医疗监护和战场侦察等领域。节点通常使用电池供电,自身的电量受限,节点通常被安放在一些危险复杂,人工难以接近的环境中,并且需要进行长时间地工作。当电池的电量被用完时,节点就会死亡从整个网络中下线。因此,节点自身的能量效率成为无线传感器网络设计中急需解决的问题。针对网络中节点能量消耗不平衡的问题,本文提出一种基于数据相似度的无线传感器网络节点调度算法,该算法主要由能耗平衡的分布式非均匀分簇算法和基于数据相似度的节点休眠调度策略两部分构成。本文的主要内容如下:首先,提出了一种能耗平衡的分布式非均匀分簇算法。针对网络中不同地域位置的节点能量消耗不平衡的问题,算法为不同区域位置的节点设置相应的竞争半径,处于不同区域的节点需要根据节点到Sink的距离、自身当前的能量水平以及邻居节点的密度来设置自身相应的竞争半径。在节点的竞争半径范围内,每个节点比较自身与其邻居节点的簇头竞选权值,选择最大权值对应的节点成为最终的簇头,普通节点根据收到的邻居节点的竞选权值情况,加入到权值最大的节点形成的簇中。其次,在上述分簇的基础上,提出一种基于数据相似度的节点休眠调度策略。根据不同数据收集子周期簇内活动节点采集的数据计算节点间的数据相似度构造模糊等价矩阵,基于邻居感知节点和数据相似度阈值的双重聚类标准进行聚类分析得到冗余节点集合。在避免感知盲区的条件下,针对冗余节点的不同聚类情况综合考虑节点间的相对位置及剩余能量进一步筛选出休眠节点,将不同子周期得到的休眠节点集合取并集操作得到最终休眠节点进行休眠调度。该休眠策略有效减少了冗余数据的收集,降低了节点的能耗。最后,使用OMNeT++仿真软件对所提的算法进行了实验测试。通过在网络能效性、网络负载均衡及数据准确度方面与其它休眠调度算法进行对比分析。实验结果显示,本文所提算法在降低节点能量消耗以及延长网络寿命方面具有良好的表现。更多还原

【Abstract】 Wireless sensor networks use nodes placed in the monitoring environment to monitor the target information required by users in different regions,and finally feedback the required data information to users.It is currently widely used in medical monitoring and battlefield reconnaissance.The nodes are usually powered by batteries,and their own power is limited.They are usually placed in dangerous and complex environments that are difficult to access by humans,and need to work for a long time.When the battery is used up,the node will die and go offline from the entire network.Therefore,the energy efficiency of the node itself has become an urgent problem in the design of wireless sensor networks.In order to solve the problem of imbalanced energy consumption of nodes in the network,this paper proposes a wireless sensor network node scheduling algorithm based on data similarity.This algorithm mainly consists of a distributed nonuniform clustering algorithm with balanced energy consumption and nodes based on data similarity.The sleep scheduling strategy consists of two parts.The main content of this article is as follows:Firstly,a distributed non-uniform clustering algorithm with balanced energy consumption is proposed.Aiming at the problem of imbalanced energy consumption of nodes in different geographical locations in the network,the algorithm sets the corresponding competition radius for nodes in different geographical locations.Nodes in different geographical locations need to be based on the distance from the node to the sink,their current energy level,and neighbor nodes.To set its own corresponding competition radius.Within the competition radius of each node,each node compares the cluster head election weights of itself and its neighboring nodes,and selects the node corresponding to the largest weight to become the final cluster head.Ordinary nodes are based on the received election weights of neighboring nodes.Join the cluster formed by the node with the largest weight.Secondly,based on the above clustering,a node sleep scheduling strategy based on data similarity is proposed.A fuzzy equivalence matrix is constructed based on thedata similarity between nodes calculated from the data collected by the active nodes in different data collection sub-period clusters.Clustering analysis is performed based on the dual clustering criteria of neighbor perception nodes and data similarity thresholds to obtain a redundant node set.Under the condition of avoiding the perception of blind spots,according to the different clustering conditions of redundant nodes,the relative positions between nodes and the remaining energy are comprehensively selected to further select the sleeping nodes,and the sleeping nodes set obtained in different sub-cycles are merged to obtain the final sleeping node.Do sleep scheduling.This sleep strategy effectively reduces the collection of redundant data and reduces the energy consumption of nodes.Finally,OMNeT++ software was used to test the proposed algorithm.By comparing and analyzing with other sleep scheduling algorithms in terms of network energy efficiency,network load balancing and data accuracy.According to simulation results,the proposed algorithm has excellent performance in reducing energy consumption of nodes and improving network survival time.更多还原