【作者】 尹玲；

【导师】 谢志军； 辛宇；

【作者基本信息】 宁波大学 ， 工程硕士（专业学位）， 2021， 硕士

【摘要】 近年来,无线传感器网络中关于节点的能量问题的研究层出不穷,在众多能量补充的方法中,无线充电技术在解决能量问题上有着不错的表现,采用无线充电方式为节点补充能量的网络就称为无线可充电传感器网络,目前已取得较好的研究进展,但也存在灵活性不够、实时性差的问题。移动充电技术作为高效且可控的无线充电方式,利用可移动的充电器依次移动到各个节点附近执行充电任务,有望彻底解决节点电池再充电问题。因此,为移动充电器设计一条合理的充电路径是当今的研究热点之一。本文研究了传感器网络中基于时空协作的充电路径规划问题,根据网络规模的大小分别设计了不同的充电路径规划方案。对于小规模网络,采用单个移动充电器执行充电任务;而对于大规模网络则使用多个移动充电器协同完成充电任务,并针对以上两种场景分别构建了合理的网络模型和充电模型、设计了两种基于时空协作的充电算法,有效延长了网络的生命周期。具体如下:(1)针对小规模网络中单个移动充电器的充电路径规划问题,首先根据节点距离基站的距离远近将网络划分成多个大小不等的簇;接下来提出了基于非均匀分簇的实时充电算法:运用动态规划算法得到最短移动路径方案,并结合节点的时、空特征确定最佳的充电顺序。实验结果表明所设计算法相较于传统按需充电算法在节点存活率和平均充电时延方面均具有较好的表现。(2)由于大规模网络场景中节点数量众多,单个移动充电器难以满足全网节点的实时充电需求,因此本文为大规模网络的充电需求设计了多个移动充电器协作充电的方案,建立了多目标优化模型,首先使用改进的聚类算法公平地划分网络为多个簇,每个簇内放置一个移动充电器;接下来研究单个簇内多目标的最优实现,提出基于时空协作的实时充电算法:结合节点的时、空特征决策充电路径。实验结果表明所提算法比现有充电算法具有更高的节点存活率和能量利用率,在其他性能指标上也有较好的表现。(3)为验证大规模网络中采用多个移动充电器协作充电的高效性,本文还运用了排队论分析所提算法的性能,结合具体场景推导理论计算公式,并将计算结果与实验结果对比,证明了本文所提的基于时空协作的多移动充电器实时充电算法的有效性。综上所述,本文研究了传感器网络中基于时空协作的充电路径规划问题,并根据网络规模的大小,分别提出了不同的充电路径规划方案,有效提高了节点存活率、延长了网络生命周期,进一步推进了传感器网络的广泛部署。更多还原

【Abstract】 In recent years,there have been endless researches on energy issues of nodes in Wireless Sensor Networks.Among many energy supplement methods,wireless charging technology has a good performance in solving energy problems,networks that use wireless charging to supplement energy for nodes are called Wireless Rechargeable Sensor Networks.At present,good research progress has been made,but there are also problems of insufficient flexibility and poor real-time performance.As an efficient and controllable wireless charging method,mobile charging technology uses mobile charging to move to each node in turn to perform charging tasks,which is expected to completely solve the charging problem.Therefore,designing a reasonable charging path for mobile chargers is one of the current research hotspots.This paper studies the charging path planning based on time-space cooperation in sensor networks,and designs different charging path planning schemes according to the size of the networks.For small-scale networks,a single mobile charger is used to perform the charging task;For large-scale networks,multiple mobile chargers are used to complete the charging task together,For the above two scenarios,reasonable network models and charging models were constructed,and two charging algorithms based on time-space cooperation were designed,which effectively extended the life cycle of the networks.Details as follows:(1)Aiming at the charging path planning problem of small-scale networks with a single mobile charger,first,the network is divided into multiple clusters of varying sizes according to the distance between the nodes and the base station;Next,a real-time charging algorithm based on non-uniform clustering is proposed: the shortest path planning scheme is obtained by the dynamic programming algorithm,and combine the time and space characteristics of the nodes to determine the best charging sequence.The experimental results show that the designed algorithm has better performance in terms of node survival rate and average charging delay than the traditional on-demand charging algorithm.(2)Due to the large number of nodes in large-scale network scenarios,it is difficult for a single mobile charger to meet the real-time charging needs of the entire network.Therefore,this paper designs a cooperative charging scheme for multiple mobile chargers for the charging needs of large-scale networks,and establishes a multi-objective optimization,the model first uses an improved clustering algorithm to fairly divide the network into multiple clusters,with a mobile charger placed in each cluster;Next,it studies the optimal realization of multiple goals in a single cluster,and proposes a real-time charging algorithm based on spatio-temporal cooperation:combine the time and space characteristics of the nodes to determine the charging path.The experimental results show that the proposed algorithm has higher node survival rate and energy utilization than the existing charging algorithm,and also has better performance in other performance indicators.(3)In order to verify the efficiency of collaborative charging with multiple mobile chargers in large-scale networks,this paper also uses queuing theory to analyze the performance of the proposed algorithm,combines specific scenarios to derive theoretical calculation formulas,and compares the calculation results with the experimental results to prove the validity of the multiple mobile chargers real-time charging algorithm based on time-space cooperation proposed in this paper.In summary,this paper studies the charging path planning problem based on spatio-temporal cooperation in sensor networks,and proposes different charging path planning schemes according to the size of the network,which effectively improves the survival rate of nodes and prolongs the life cycle of the network,further promoted the widespread deployment of sensor networks.更多还原