【作者】 宋永献；

【导师】 张荣标；

【作者基本信息】 江苏大学 ， 控制理论与控制工程， 2014， 博士

【摘要】 随着无线传感器网络(WSN)不断发展,其应用领域越来越广泛,而在许多应用场合对WSN数据可靠传输要求很高。特别在一些恶劣环境下,无线传感器网络易受现场环境、衰减、多径、盲区以及节点性能等不利因素影响,数据传输时容易产生错误和丢包,数据可靠传输得不到保障。因此,无线传感器网络在数据可靠传输方面研究面临着诸多挑战。在分析和总结WSN数据可靠性传输研究现状、相关研究领域以及采取研究方法的基础上,针对无线传感器网络可靠数据传输研究的不足之处,提出相应研究目标和思路,并取得相关研究成果如下：1、基于能量均衡的数据可靠传输研究。无线传感器网络由于能量负载分布不均衡,一些节点由于能量过度消耗过早死亡,造成通信链路中断以及数据丢包,从而影响数据可靠传输。在分析影响WSN能耗的主要因素和现有节能技术基础上,结合虚拟多输入多输出(Virtual MIMO)路由算法在同构无线传感器网络应用情况,提出一种应用于中小规模同构WSN的虚拟MIMO分簇(VMMCA)算法。该算法不仅能够实现簇头节点循环随机选取,而且还能够在保证节点通信质量前提下实现WSN生命周期优化。建立一种基于虚拟MIMO的分簇网络能耗模型,在不同簇域大小、节点分布密度和路径损耗指数以及汇聚节点不断变化的情况下,分析虚拟MIMO网络与SISO网络的能耗变化规律。为了能够使网络生命周期达到最大化,网络能量达到均衡,应用遗传算法对簇头比例进行优化。实验仿真研究表明,与LEACH算法相比,VMMCA算法能够很好实现能量分布均衡,延长了网络生命周期。2、基于多维高斯自回归模型的无线传感器网络链路度量方法。为了能够满足检测事件需求信息的可靠性,以达到网路生命周期最大化,提出一种基于多维高斯自回归模型的协作路由策略。该策略通过多维高斯自回归模型分析检测领域内信息相关性,并对路径每一个链路的参与进行量化,以减少在融合中心做最终决策时的误差概率。传感器节点传输到融合中心的感知信息中仅仅包括新信息而不是原始信息,并采用快速滤波器对沿着到融合中心路径上的中间节点聚集数据信息进行滤波。链路表达式中引入了能耗加权值以实现协作路由策略中检测可靠性和能耗性平衡,并且分析加权因子ω、相关阶数m和网络规模变化对协作路由性能影响。实验仿真结果表明,在给定检测可靠性标准下,与非协作路由策略以及传统最小能量路由策略相比,提出的协作路由策略在能效和检测可靠性方面都得到了明显改善。3、基于链路指示的WSN可靠数据传输研究。通过实验验证天线方向、通信距离、发射功率、节点电压以及外界动态环境对LQI均值和标准差率CV的影响,在此基础上,构造出一种基于链路质量指示(LQI)均值和标准差率CV的路由度量模型(Routing Metric Based on Link Quality Indication and Coefficient of Variance, RMBLQICV)。为了能够有效评价数据传输优劣,选出性能最佳传输路径,RMBLQICV模型把LQI均值与标准差率实现动态结合,瓶颈链路出现概率被有效降低,数据可靠传输获得有效提高。针对经典按需路由协议的不足,对其进行了改进,并结合提出的RMBLQICV模型,获得一种基于LQI均值的多路径AODV路由协议(LBM_AODV)。通过由CC2430作为接收模块的实验平台对RMBLQICV模型和LBM_AODV协议进行实验验证。结果表明,在路径选择上,相对LQI均值模型,RMBLQICV路由度量模型有效避免误判情况出现,准确剔除含有瓶颈链路的路径,数据可靠传输得到提高。4、基于链路感知的三维WSN可靠数据传输研究。为了描述空中无线传感器网络通信链路行为,采用基于CC2430的无线传感器信息传输平台设计一系列实验来详细观察各种因素对WSN链路性能影响。实验结果表明,除了通常户外环境因素对链路性能影响之外,天线方向和由于地面反射产生的多径衰落是影响空中WSN链路性能衰减的重要因素。在充分考虑影响链路性能衰减因素的基础上,提出一种基于链路感知WSN数据可靠传输协议(Data Reliable Transmission Protocol Based on Link Aware in WSN, DRTPLA_WSN),详细分析了该协议的设计和性能评估,分别在具有容忍延迟和实时性要求的两种AWSN (Aerial WSN)环境下对DRTPLA_WSN进行评估。仿真结果表明,DRTPLA WSN有效改善了网络整体性能,对于实时传输的空中无线传感器网络,其丢包率大幅度下降,大大提高了网络可靠性,节约了网络能耗。更多还原

【Abstract】 With the continuous development of wireless sensor network, its applications become more and more widely. However, data transmission of WSN requires high reliability in many applications. In some particularly harsh environments, wireless sensor network is often affected by complex environment, attenuation, multi-path, dead-zone, node performance and other factors, which will easily lead to errors and data packet loss in data transmission process, data reliable transmission can not be guaranteed. Therefore, reliable data transmission has become a significant problem in the research of wireless sensor network.On the basis of analyzing and summarizing data reliable transmission research status, related research fields and research methods for WSN, focusing on the lack of reliable data transmission research in wireless sensor network, the research goal and outline are proposed, and related research results are as follows:First, study on data reliable transmission based on energy balance in WSN. In view of energy load uneven distribution of WSN, some nodes will die prematurely due to excessive energy consumption, which lead to interruption of communication links and data packet loss, thus reliable data transmission is affected. On the basis of analyzing the main factors that affect energy consumption and existing energy saving technologies, combined with application of virtual multiple-input multiple-output (Virtual MIMO) routing algorithm in isomorphic wireless sensor network, virtual multiple-input multiple-output clustering algorithm (VMMCA) which applies to small and medium scale isomorphic WSN is proposed. VMMCA not only can select cluster head randomly, but also can achieve the life cycle optimization of WSN on the premise of assuring nodes communication quality. Virtual MIMO cluster network energy consumption model is established. On the condition of changing for different clusters size, node distribution density, the path loss index and sink nodes, the change of the energy consumption of virtual MIMO network and SISO network is analyzed. In order to balance network energy load and prolong lifetime of WSN, and the network lifetime is taken as the optimization target, the ratio of the clusters head is optimized by genetic algorithm. The experiment and simulation results show that compared with LEACH algorithm, VMMCA can achieve very good balance of energy and prolong network lifetime.Second, links metric based on multi-dimensional Gaussian autoregressive model in WSN. In order to meet the demand of detection events information reliability and maximize network life cycle, the cooperative routing strategy based on multi-dimensional Gaussian autoregressive model is proposed. The cooperative routing strategy adopts multi-dimensional Gaussian autoregressive model to analyze information correlation within the detection field, and the participation of each link on a path is quantized to reduce the error probability when making final decision at the fusion center. Only new information instead of whole original data is included when sensed information is propagated to fusion center by sensor nodes, and using the quick filter to aggregate data information at intermediate nodes along the path to the fusion center. In order to balance detection reliability and energy efficience, and energy consumption weighted value is introduced in the expression of the detection reliability-aware link metric. Simulation results show that at a given detection reliability standards, compared with non-cooperative routing strategies and traditional minimum energy routing policy, the proposed cooperative routing strategy has been significantly improved in terms of energy efficiency and reliability of detection.Third, study on data reliable transmission based on link quality indication in wireless sensor network. The influence of antenna orientation, communication distance, transmission power, supply voltage and external dynamic environment on LQI mean and Coefficient Variance (CV) is verified by experiment, and on the basis of it, a kind of routing metric model based on LQI mean and CV (RMBLQICV) is constructed. In order to effectively evaluate the merits of data transmission, choose the best transmission path, and RMBLQICV model dynamically combines LQI mean and CV to evaluate the quality of paths, thus it effectively reduces the appearing likelihood of bottleneck link and improves reliability of data transmission. Aiming at lack of a classic on-demand routing protocol, the on-demand routing protocol AODV is improved, and combine proposed RMBLQICV to obtain Multipath Routing Protocol (LBM_AODV) based on mean LQI. RMBLQICV model and LBM_AODV protocol are verified through the experimental platform which takes CC2430as transceivers, the experiment results show that compared with LQI mean model, RMBLQICV routing metric model effectively avoid misjudgment in terms of choice of the path, and accurately eliminate the paths which contain bottleneck link, data reliable transmission is greatly improved.Fourth, study on data reliable transmission based on link aware in three-dimensional wireless sensor network. In order to describe behavior of air wireless sensor network communication links, the performance and reliable data transmission of links in the outdoor environment are studied. A series of experiments based on CC2430wireless sensor platform are designed to observe various factors that affect performance of AWSN in detail. The experiment results show that, in addition to usual outdoor environmental factors affecting the link performance, two important factors to the link degradation in aerial wireless sensor networks are direction of antenna and multipath fading due to reflections of the ground. On the basis of fully considering factors affecting performance degradation of link, a kind of data reliable transmission protocol based on link aware in WSN (DRTPLA_WSN) is proposed, and the agreement design and performance assessment are analyzed in detail, DRTPLA_WSN is evaluated in tolerate delay and real-time requirements AWSN environment respectively. Simulation results show that, DRTPLA_WSN effectively improve the overall performance of network, packet loss rate decrease significantly for real-time transmission air wireless sensor networks, which greatly improves the reliability of network and saves energy consumption of network.更多还原