基于分簇结构的无线传感器网络节能路由协议研究

发布时间：2018-03-11 01:15

本文选题：无线传感器网络　切入点：分簇路由　出处：《重庆大学》2016年博士论文　论文类型：学位论文

【摘要】：典型的无线传感器网络(Wireless Sensor Network,WSN)是由许多微型廉价低功耗的传感器节点采用自组织的方式形成的专用网络,常被部署于指定的区域用于感知和采集各种监测对象的信息,并采用多跳通信方式将信息提供给终端用户,是连接人类社会和物理世界的信息纽带。WSN与传统的无线网络不同,WSN中的节点带宽、内存等资源更为匮乏,尤其是其有限的能量资源直接影响到传感器网络的生命周期以及网络的信息质量。由于传感器节点的能量通常很难得到补充,因此,在节点能量有限的前提下,讨论如何提高有限能量的利用率,以延长网络的生命周期,是当前该领域研究的热点问题。传感器节点绝大部分的能量都用于数据的无线传输,因此采用高能效的路由协议有助于减少节点的能耗,从而延长网络生存期。通过对现有的路由协议系统学习和总结,发现基于分簇结构的路由协议在网络扩展性、高效节能性等方面具有明显优势,本文围绕这一课题针对不同类型的WSN构建分簇结构的路由协议。论文主要内容及创新性成果包含:(1)针对传统同构的WSN,研究如何构建单跳分簇路由协议,提出了一种混合的基于博弈理论的分布式分簇路由协议(Hybrid,Game Theory based and Distributed clustering,HGTD)。在传统WSN单跳分簇路由协议中,簇头节点采用单跳通信的方式将数据传输至较远的基站,能耗速率很快,如何选择最优的簇头节点以均衡节点间的能耗并延长网络生存期是一个研究重点。本协议中引入博弈理论用于解决单跳分簇路由中的簇头节点选择问题,节点被建模为博弈参与者,通过与其邻近区域的节点博弈得到自己成为簇头的均衡概率。在分簇博弈模型中特别定义了节点选择不同策略时的收益,并考虑到节点度以及节点到基站的距离,使得节点在根据均衡概率决定是否成为簇头时能够在最小化能量开销和提供必须的网络服务之间保持均衡。此外,设计了一种迭代算法从潜在的簇头节点中选出最终的簇头节点,可均衡节点间的能耗并且避免相邻的节点同时被选为簇头。仿真结果表明本协议能够明显改善网络生存期。(2)由于外界因素的影响,传感器节点被部署于监测区域一段时间后通常能量异构。本文针对能量异构WSN,研究如何构建多跳分簇路由协议,提出了一种基于非均匀簇结构的路由协议(Unequal Cluster-based Routing scheme for multi-level Heterogeneous wireless sensor networks,UCR-H)。在多跳分簇路由协议中,网络中的所有簇头节点共同合作将数据以多跳通信方式传输至基站,因此离基站越近的簇头节点担负的簇间数据转发任务越重,这会导致能量空洞问题的发生。为了避免在能量异构网络中采用多跳分簇路由协议时出现能量空洞问题,本协议中将网络所在监测区域划分为若干等尺寸的矩形单元。首先通过均衡不同单元间簇头节点的能耗,计算得到各单元中的簇头数目;然后通过最小化网络中簇间数据转发的总能耗,得到最优的单元数;最后根据节点的能量等级以及节点所在单元的簇头数目,为各单元中不同能量等级的节点设计了当选为簇头时对应的簇半径,此外还得到了一轮次阈值避免对能量等级高的节点的过度惩罚。仿真结果表明UCR-H协议能够有效减轻网络中的能量空洞问题,并且明显改善了网络生存期。(3)近来,在一些WSN应用中传感器节点配备有能量获取装置,但从环境中获取能量的能力很有限。本文针对能量获取WSN,研究如何构建多跳分簇路由协议,提出了一种多跳的能量中性分簇路由协议(Multi-hop Energy Neutral Clustering,MENC)。设计本协议的主要目的是实现对节点从环境中获取的能量的有效利用,以获得无限长的网络生存期。通过结合节点的能耗情况以及能量状态,给出了节点的能量中性约束,在此约束下节点可工作在能量中性状态下,从而可维持网络的永久运行,避免网络失效。此外,在确保节点工作在能量中性状态的前提下,通过凸优化的方法得出最小的网络数据传输周期,使得基站成功接收到的数据量达到最大值。仿真结果表明MENC协议能够获得无限长的网络生存期,确保网络中节点持续的数据采集与传输,并且网络吞吐量得到明显提高。
[Abstract]:A typical wireless sensor network (Wireless Sensor Network, WSN) is a special network formed by many cheap miniature low power consumption of the sensor nodes in a self-organized way, are often deployed in the designated area for sensing and gathering information on a variety of monitoring objects, and adopt multi hop communication to provide information to the user terminal, is different information link.WSN with traditional wireless network connection of human society and the physical world, the bandwidth of WSN, memory and other resources more scarce, especially the limited energy resources directly affects the quality of information sensor network life cycle and network. As the supplement, the energy of sensor nodes are usually difficult to get it in the premise of limited node energy, discuss how to improve the utilization rate of energy limited, to extend the network life cycle, is the focus of research in this field. Most of the energy of sensor nodes are used for wireless transmission of data, so the use of energy-efficient routing protocol helps to reduce the energy consumption of the nodes, prolong the network survival. By summarizing the existing system of routing protocol and routing protocol that cluster on scalability in network based on high performance has obvious advantages such aspects, this paper focuses on the subject for different types of WSN to construct the routing protocol clustering structure. The main contents and innovative achievements include: (1) according to the traditional homogeneous WSN, studied how to construct the single hop clustering routing protocol, we propose a hybrid distributed clustering routing game theory protocol based on (Hybrid, Game Theory based and Distributed clustering, HGTD WSN). In the traditional single hop clustering routing protocol, cluster head nodes using single hop communication mode for data transmission to the distant base Station, energy consumption rate is fast, how to select the optimal cluster head nodes to balance the energy consumption of nodes and prolong the network lifetime is a focus of research. The game theory is introduced in this agreement to solve the single hop clustering routing in cluster head node selection problem, the node is modeled as game participants, nodes of the game and the adjacent area of the equilibrium probability become cluster head. In the cluster game model defines the special node selection strategy of different benefits, and considering the node degree and node distance to the base station, the node can keep balance between minimizing energy cost and provide the necessary services in the network according to the equilibrium probability to decide whether as the cluster head. In addition, the design of an iterative algorithm from the cluster head node potential in the selected cluster head node finally, can balance the consumption of the node and avoid adjacent nodes at the same time Was selected as the cluster head. The simulation results show that this protocol can improve the network lifetime. (2) due to the influence of external factors, the sensor nodes are deployed in the monitoring area after a period of time. According to the usual energy heterogeneous energy heterogeneous WSN, studied how to construct the multi hop clustering protocol is proposed, which is based on non uniform routing protocol cluster structure (Unequal Cluster-based Routing scheme for multi-level Heterogeneous wireless sensor networks, UCR-H). In the multi hop routing protocol, all cluster nodes in the network cooperate data in multi hop communication mode of transmission to the base station, the cluster head nodes closer to the base station is responsible for the inter cluster data forwarding the heavier tasks, this will occur. In order to avoid the energy hole problem in energy heterogeneous network the energy hole problem of multi hop clustering protocol, this protocol The network where the monitoring area is divided into several rectangular element size. Firstly, the balance of energy consumption between different elements of the cluster node, calculate the number of cluster head of each unit in the network; then by minimizing the total energy consumption of the inter cluster data forwarding, get the optimal number of units; finally, according to the number of cluster head node energy level and the node where the unit, for different energy levels of nodes each unit is designed to be elected as the cluster radius corresponding to the cluster head, in addition to get a round threshold to avoid excessive punishment of the node energy level high. The simulation results show that UCR-H protocol can effectively alleviate the problem of the network energy hole, and significantly improved the survival of the network. (3) recently, in the application of WSN in some sensor nodes are equipped with energy harvesting device, but the ability of obtaining energy from the environment is limited. According to the energy received Take WSN, study how to build a multi hop clustering protocol, proposed an energy neutral multi hop clustering routing protocol (Multi-hop Energy Neutral Clustering, MENC). The main purpose of this agreement is designed to achieve efficient use of nodes to obtain energy from the environment, in order to obtain the network lifetime is infinite. By combining the energy consumption of the node energy state and, given the node energy neutral constraint, this constraint node can work in energy neutral condition, which can maintain a permanent operation of the network, to avoid network failure. In addition, in the premise of ensuring the node work in energy neutral state, network data obtained by the minimum transmission cycle a convex optimization model, the amount of data the base station receives to reach the maximum value. The simulation results show that the MENC protocol can achieve network lifetime infinite, ensure that the nodes in the network to Continuous data acquisition and transmission, and network throughput has been significantly improved.

【学位授予单位】：重庆大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TN929.5;TP212.9

【参考文献】