【作者】 邓俊；

【导师】 吴玉成；

【作者基本信息】 重庆大学 ， 通信与信息系统， 2011， 硕士

【摘要】 无线传感器网络(Wireless Sensor Networks,WSN)由大量廉价的传感器节点所构成,节点间通过无线通信方式形成一个多跳自组织网络。普通传感器节点可以通过其携带的各种类型的传感器采集现场的数据,并以接力的方式通过其它传感器节点将采集到的数据逐跳地发送到汇聚节点(Sink)。由于无线传感器网络节点资源受限,如何通过拓扑管理、能量有效的路由协议、功率控制等方面来延长网络的生存时间是无线传感器网络的核心研究课题。无线传感器网络路由协议为节点数据的传输指明路径,它影响和决定了整个无线传感器网络的性能和寿命。传输路径上的中间节点通过数据融合技术对来自不同节点的数据进行处理,从而减少传感器节点间数据通信,是节省传感器节点能量开销,延长网络寿命的有效技术手段之一。论文主要研究利用数据融合技术解决无线传感器网络由于能量限制导致的生存性问题,并探讨基于数据融合的无线传感器网络路由协议在网络生存时间方面的性能。在无线传感器网络中,影响路由的因素包括业务类型、拓扑结构、节点处理能力等各个方面,针对已有的无线传感器网络路由算法极少考虑多种干扰因素的问题,又考虑到几何路由算法是对数据融合支撑性最好的一类路由算法,论文提出了一种基于势能场的几何路由算法,通过距离势能场把二维的无线传感器网络转化到三维空间讨论,并且通过场的概念,使无线传感器网络路由算法可以通过引入不同的势能场的叠加来解决多个因素对路由的影响。仿真分析表明,此算法可以有效地处理多种因素对路由效果的影响,并且能快速地收敛。论文在引入能量消耗模型的基础上,研究数据融合与路由对网络生存时间的影响。由推导得出,基于数据融合与路由的联合优化,网络生存时间的最大化问题可转化为一个不可微的最优化问题。引入瓶颈节点的概念,利用光滑函数近似的方法,把该最优化问题转化为二阶函数的极大值问题,并得出了有解的充分条件。利用此充分条件和几何路由协议的思想,论文提出了一种分布式路由算法并进行了仿真分析。结果显示,此算法能显著减少数据流量和增大网络生存时间,同时能有效快速的收敛。更多还原

【Abstract】 Wireless Sensor Network (WSN) is a multi-hop ad hoc network which consists of a large number of spatially distributed autonomous sensors to monitor physical or environmental conditions. Ordinary sensor nodes can carry many types of sensors for field data collection, and send collected data to the sink node hop by hop.Since wireless sensor network node is resource-constrained, how we utilize the technology of topology management, energy efficient routing protocols and power control to extend the network lifetime is the core of the wireless sensor network research. Wireless sensor network routing protocol specifies path for the data transmission, which influences and determines the performance and lifetime of wireless sensor network. Intermediate nodes on the transmission path can reduce the data communication among sensor nodes through data aggregation, thereby extending the network lifetime. The thesis makes use of data aggregation technology to solve the survival isssues caused by energy constraints of wireless sensor network, and explores the network’s lifetime performance with the data aggregation-based routing protocol. In wireless networks, the factors which impact routing are service type, topology, node process ability and other aspects. The existing routing algorithms for wireless sensor networks rarely consider a variety of interfering factors, and taking account of that the geometric routing is the best support for the data aggregation in routing protocols, the thesis proposes a potential field-based geometric routing algorithm, transforms the two-dimensional wereless sensor network into three-dimensional network through distance potential field, and addresses the impact of multiple factors on the route by the superposition of different potential fields. Simulation analysis shows that this algorithm can effectively handle a variety of factors which affect the routing path. And this algorithm has a quick convergence.The thesis introduces a energy consumption model to study the influence of the network lifetime by the data aggregation and routing protocol. Obtained by the derivation, network lifetime maximization problem can be transformed into a non-differential optimization problem based on jointly optimizing data aggregation and routing. The thesis also introduces the concept of bottleneck nodes and uses a smooth function approximation method to transform the optimization problem into the maximum value problem of second-order function, and also obtains sufficient conditions when the function is solvable. The thesis proposes a distributed routing algorithm based on the sufficient conditions and geometric routing protocols. Simulation analysis shows that this algorithm can handle a variety of factors which affect the routing path. And this algorithm has a quick convergence.更多还原