多Sink环境下的无线传感器网络数据汇集问题的研究

发布时间：2018-03-25 13:24

本文选题：多Sink无线传感器网络　切入点：数据汇集　出处：《福州大学》2014年硕士论文

【摘要】：集成了数字电子技术、微机电系统和无线通信技术而形成的无线传感器网络是一种获取信息和处理信息的新技术。无线传感器网络是由具有感知能力、计算能力和通信能力的微型传感器节点所组成,在军事国防、环境监控和医疗健康等领域展现了广阔的应用前景。随着传感器网络规模的扩大,多个Sink节点环境下的研究越来越得到重视。无线传感器网络中节点的密集分布,使得相邻节点感知到的数据存在相关性。为减少数据的冗余传输,网内数据汇集被提出并成为无线传感器网络的一项重要技术。在实时的应用中,为保证数据汇集的时效性,端到端时延是一项重要的性能指标。现有基于树形结构的汇集调度都是按照先构造汇集树,再为节点分配时隙这两阶段来实现,并不能充分利用时隙,导致汇集时延较大。此外,现有的大部分汇集工作都是针对可靠通信环境展开,在更符合实际的不可靠通信环境下研究能量有效的数据汇集在能量受限的传感器网络中显得更有意义。基于上述考虑,本文研究了最小时延数据汇集调度问题和不可靠通信环境下最大化网络生命期的数据汇集问题。(1)研究了最小时延数据汇集调度问题,提出了分层的汇集调度算法,能够有效降低网络的数据汇集时延。本文利用广度优先搜索对传感器网络的节点进行分层。然后,文中根据分配的时隙序号和该时隙的最大化利用逐层地为节点选择父节点。基于此,本文分别提出了基于极大独立集和不采用独立集以进一步扩展父节点选择范围的汇集调度算法,并从理论上对算法的性能进行分析。实验结果表明,本文所提出的算法能够充分利用时隙,以低时延进行汇集调度。(2)研究了不可靠通信环境下最大化网络生命期的数据汇集问题,提出树形结构的汇集算法和有向无环图的汇集算法,能够有效延长网络生命期。本文首先分析了相邻节点之间的通信,给出了为保证节点间成功通信所消耗的平均能量最小时的最优发送半径,并分析了如何为传感器节点选择Sink节点的问题。在此基础上,本文以均衡节点的能量消耗为导向,依据节点的剩余能量和通信代价,提出了基于树形拓扑的汇集算法,以及可以提高数据接收概率的有向无环图汇集算法。实验结果表明,本文提出的算法在网络生命期上能取得较优的性能。
[Abstract]:The integration of digital electronic technology, wireless sensor network, MEMS and wireless communication technology which is a new technology of acquiring and processing information. Wireless sensor network is composed of micro sensor nodes with sensing capability, computation and communication capabilities of the group, military defense, environmental monitoring and medical and health fields show a broad application prospects. With the expansion of the scale of sensor networks, multiple Sink nodes environment gets more and more attention. The dense distribution of nodes in wireless sensor networks, a correlation exists between the adjacent nodes to sense data. In order to reduce the data redundancy, data collection is proposed and become an important technology in wireless sensor networks. In real-time applications, in order to ensure the timeliness of data collection, the end-to-end delay is an important performance index of the existing base. In the aggregation tree structure are first constructed in accordance with the collection of trees, to the two stage to assign time slots to nodes to achieve, and can not make full use of time slots, resulting in pooled delay. In addition, most of the existing collection of work are based on the reliable communication environment, with unreliable communication environment in the practical research data energy efficient collection of more significance in energy constrained sensor networks. Based on the above considerations, this paper studies the minimum delay of data collection scheduling problem and unreliable number to maximize the network lifetime communication environment according to the collection of problems. (1) the minimum delay of data collection scheduling problem, put forward a hierarchical collection scheduling algorithm that can effectively reduce the network delay of data collection. This paper uses the breadth first search nodes to the sensor network layer. Then, according to the assigned time slot sequence Layer select the parent node to the node with the maximum number and time slot. Based on this, this paper proposes a maximal independent set and the independent set to further expand the scope of the collection of the parent node selection algorithm based on scheduling, and analyze the performance of the algorithm theoretically. Experimental results show that the proposed the algorithm can make full use of time slots, pooled scheduling with low delay. (2) study is not reliable to maximize the network lifetime data communication environment together, put together the tree structure and algorithm of acyclic graph collection algorithm, can effectively prolong the life time of the network. This paper analyzes the communication between adjacent nodes the proposed to ensure the inter node average energy consumption of the most successful communication hours of optimal transmission radius, and analyzes how to choose Sink sensor node node problem. On the basis of In this paper, in order to balance the node energy consumption oriented, based on the residual energy of nodes and the communication cost, is proposed based on tree topology aggregation algorithm, and can improve the data receiving probability of the directed acyclic graph collection algorithm. The experimental results show that the proposed algorithm can achieve better performance on Network lifetime.

【学位授予单位】：福州大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN929.5;TP212.9

【相似文献】