【作者】 孙亭；

【导师】 杨永田；

【作者基本信息】 哈尔滨工程大学 ， 计算机应用技术， 2007， 博士

【摘要】 近年来,微机电系统(MEMS)、低功耗高集成度电子器件及无线通信技术的快速发展,导致低成本、微体积、多功能的无线传感器节点设备的出现。在未来几年内,无线传感器网络(WSN)将对几乎所有的工业领域和人们的日常生活带来巨大影响。然而,由于无线传感器网络节点具有电池供电、不可回收等特点,其节点能量有限、存储能力很小、传输距离有限,能量问题成为影响无线传感器网络性能的关键问题。路由机制从网络层决定整个网络的工作状态,简单、有效、低功耗的路由成为高性能网络的重要指标。以能量优化为目标,本文主要研究了基于簇的无线传感器网络分层路由技术,其中包括簇内节点的数据传输、簇的动态生成过程、簇间信息交互以及基于应用的网内路由解决方案。首先,本文提出了一种簇内基于最小跳数的路由算法。网络总体采用基于簇的分层结构,然后在簇内采用基于最小跳数的路由模式进行数据采集与传输。这种结构充分发挥了簇内节点数量较多时,最小跳数路由的优势,可以保证任何节点发出的信息都沿着最优的路径向簇首传输,并且在网络内所引起的信息包数量最少。所有节点只要记忆自己的最小跳数和一跳范围内的转发节点集,就可以实现信息路由。新加入的节点只要向自己周围的一跳邻居发出注册信息即可加入网络,从而达到了提高网络性能、降低网络能耗的目的。接着,本文提出了一个基于节点聚合度的动态分簇路由算法。它以节点的聚合度为依据,将网内节点分簇,利用聚合节点间关系紧密的特点,降低网络频繁整体分簇的管理能耗。这样建立的簇内节点互相联系密切,同时降低了簇之间的节点重叠。在簇内,采用动态分簇的理念,结合簇重组和簇的自愈机制,使得路由协议具有动态平衡节点能耗的特点。簇首间采用多跳的数据传输模式,降低了传输能耗,从而达到延长网络寿命的目的。针对按需网络环境的特点,本文提出了一种簇中建链的路由算法。在按需网络环境下,基于簇的分层路由传输模式在一定程度上减轻了网中节点能量的集中消耗,但是簇内节点通信过程中的功耗均衡问题是个大问题。链式通信通过将簇内节点组链,使得数据在从源节点向簇首的传输过程中就得到了充分的融合,降低了传输能耗。链首作为簇首的代理节点,分担了簇首的工作,降低了簇首的能耗,而链首的动态更换使得节点间的能耗更加平均。针对无线传感器网络与实际应用相结合的特点,本文提出了一种能够解决网内可移动节点运动过程中与静态节点顺利通信问题的移动节点路由算法。把基于移动代理的信息处理过程扩展到簇内,将移动节点在簇内沿最优路径收集数据的问题等价成一个顶点加权的游客问题,通过采用一种新的近似算法,由簇首计算出移动节点在簇内运动的最佳路径,使移动节点能够采集到最充分的静态节点数据。新路由算法给出了各要素彼此之间的关系,保证了移动代理节点对固定静态源节点的信息的成功接收,与此同时结合动态休眠机制,降低了节点能耗。全文的工作围绕分层的无线传感器网络路由展开,结合无线传感器网络的特点及其实际应用环境的特点,来解决目前领域内存在的问题。更多还原

【Abstract】 In recent years, some new low cost and multi-function micro-sensor nodes are produced along with the rapid development of Micro-Electro-Mechanical Systems, low energy-consumption electro-devices and wireless communication techniques. And in the future, wireless sensor networks that consist of these equipments will probably bring huge influence to almost all of the industrial fields and people’s daily life. However, as wireless sensor nodes are characterized by limited energy availability, low memory and reduced processing power, energy efficiency is a key issue in designing the network. Routing mechanism can affect the network’s status from the network layer; simple, effective and low energy consumption route becomes an important target of a high performance network. Taking the energy optimization as the goal, this dissertation mainly researches on the cluster-based hierarchical routing protocols, including the communication between nodes in cluster, dynamic foundation of the cluster as well as application-based routing solution.In the dissertation, a new least-hop based routing algorithm is proposed. The network overall uses the cluster-based hierarchical routing mechanism, and in cluster, least-hop based routing model is used for data collection and transmission. This kind of structure fully utilizes the advantage of least-hop route while the nodes is more in cluster, which can make the communication between nodes along with the best route, and can make the information packets in the network least. So long as remember their own hop number and those nodes that are one-hop far away from them, the source nodes can transmit data to their cluster-heads easily. Regarding the nodes that newly join, they can become the member of the network just after they send the registration information to their one-hop neighbors. Then, it achieves the goal of enhancing the network performance and reducing the energy consumption.A new convergence-degree based hierarchical cluster routing algorithm is also proposed here. It takes the node’s convergence degree as the basis, and divides the network nodes into clusters, and fully displays the close relationship between the convergent nodes and reduces the network management energy consumption while frequently re-clustering. With this kind of structure, the nodes in same cluster will have more close relationship, and the network will have lower node overlapping between different clusters. In cluster, by using dynamic clustering organization, union with cluster rebuilding and cluster’s self-recovery mechanism, the algorithm enables the protocol to have the dynamical equilibrium energy consumption characteristic, which balances the nodes’ energy consumption. Between the cluster-heads, multi-hop routing model is used, which can reduce the transmission energy consumption, and then prolongs the lifetime of the network.In the on-demand network environment, a method that a cluster-head distributes energy load evenly among its members is proposed. Cluster-based routing protocol can reduce the nodes’ energy consumption to a certain degree, but how to balance the nodes’ energy consumption in cluster is still a major problem. By building the chain in cluster, the data can be fully fused while being translated from the source node to cluster-head, and the new route pattern needs less transmission energy. As the agent node of cluster-head, the head of chain shares the cluster-head’s work, which reduces the cluster-head’s energy consumption. At the same time, the dynamic replacement of chain-head causes the nodes energy consumption more balanced.As wireless sensor network is application-based, a mobile agent routing algorithm in cluster is also proposed, which can resolve the communication problem between mobile nodes and static nodes of the network. It regards the problem that mobile nodes collecting data in cluster along with superior way as a vertex-weighted traveling tourist problem, at the same time expands the mobile proxy based information processing process into the cluster. By using a new approximation algorithm, the cluster-head can get the best route that mobile nodes will along with, and then the mobile proxy will gather the fullest data from the static nodes. The new algorithm gives us the relationship between various essential factors, and enables the successful data transmission between nodes. And union with the dynamic dormancy mechanism, it can reduce the energy consumption.This dissertation mainly researches on the hierarchical routing algorithm of wireless sensor networks. Based on the networks’ characteristics and its practical application environments’ characteristics, the aim of the work is to resolve the present problems in domain.更多还原