【作者】 孙大洋；

【导师】 刘衍珩；

【作者基本信息】 吉林大学 ， 计算机应用技术， 2010， 博士

【摘要】 本文在详细分析了各种节能技术,对节能技术进行微观宏观分类的基础上,从软件层面对传感器网络的节能问题展开了深入的研究。主要贡献和创新点包括以下几个方面:1.提出了数据压缩以及数据融合的节能度量模型,为数据压缩与数据融合方法的选择提供了一种决策方法。并以此为基础提出了一种基于能耗度量的融合树构建算法,通过融合节点反馈能耗以及到达Sink节点的跳数信息,对多个路由的能耗进行评估,进而选择低能耗路由。2.提出了基于动态博弈理论的能量感知路由算法,将能量感知路由选择问题建模为动态博弈问题,解决了能量感知路由以能耗统计值随机选择下一跳路由节点的缺陷,提高了网络节点路由选择的可预测性,为休眠调度机制提供了有力的技术支持,该路由算法更加适合与休眠调度机制相结合从而延长无线传感器网络的生命期。3.对定位技术中减少定位信息量的方法进行了研究,提出了使网络节点定位具有确定性约束的约束唯一实现理论,降低了网络定位的能耗上限,由此建立的约束定位方法不仅提高了网络的定位精度,而且使得网络定位所需的定位信息大为减少,该方法同时降低了网络连通度的下限,减少了网络节点退避重传的几率,从而达到网络节能的效果。4.提出了系统设计全生命周期的面向服务的网络生存期优化体系模型,该模型利用先验统计信息将不同的应用与优化算法统一起来,从服务“供求”的角度对具体应用服务需求与网络服务能力进行量化评估,并以评估结果作为反馈调节网络使之达到网络生存期最大限度的优化。该模型将各种节能技术在服务的层面上综合考虑,贯穿于应用解决方案的各个环节,为传感器网络节能需求发展为应用解决方案的一项独立需求提供理论上的支持。更多还原

【Abstract】 Severe energy constraint is one of the differences between WSN and traditional network. It is not feasible to supply energy with new batteries to the sensor nodes because there are plenty of sensors which are very cheap and are deployed in a wide and complex area, some of which men can not reach. Besides, WSN can not be reused after its energy burns out because of the features of large scale and its battery not easily replaceable, which makes energy efficient technologies a method to decrease the cost of the network per hour. In the per-hour-cost point of view, energy saving are not only an issue of technology, but also an issue of cost, which can play a competitive role between companies. Even if solar power or vibration technologies can be used in the future, all the energy saved can be stored to supply a more complex application, which can make WSN spread widely and also can make companies competitive. Therefore, energy efficient technologies are very important in WSN, the principal goal of energy aware WSN is the efficient use of energy.The energy aware issue has become a very hot research point in recent years. A lot of researchers have focused on diminishing the network power consumption. Such work can be illustrated from micro and macro perspectives respectively. In micro aspect, data compression and data aggregation can be used to reduce the power consumption by transmitting less data. Meanwhile, the nodes can be arranged to get into the sleep mode when there is no transmission task, which can be called sleep scheduling. These researches from micro angle have laid a solid foundation for the power saving issue, and furthermore, sleep scheduling plays a very important role especially when the scenario has a low frequency of data transmission. But in the micro aspect, the schemes have only considered the power consumption of individual nodes or path, which has a feature of centralization, that is, the feature of burning the energy of the nodes along those paths. This leaves the network with a wide disparity in the energy levels of the nodes, eventually leading to disconnected subnets, and thus the network loses the ability to serve. However, as is known to all that the goal of the network is to gather information from a specified region of observation and relay this information to the sink node, whether the network can support service should be concentrated on, not the lifetime of a single path. This inspires the work of macro energy saving. In macro aspect, the energy aware issue has stepped into a new stage on which the aim of energy aware technologies should be more service-oriented and the lifetime should be prolonged more systematically.A whole study of energy aware issues is carried out in this paper in the software perspective view. After a detailed analysis on various energy aware technologies, the issues are deeply investigated. In micro aspect, data compression and data aggregation, as well as the scheme of sleep scheduling and energy efficient localization is explored. The research work of micro aspect can gain an energy efficient impact on local areas. However, for WSN with the aim of gathering data, the essential goal is to hold this issue as a whole. Therefore, in the macro aspect of energy saving, service-oriented lifetime is studied in this paper, treating micro-aspect technologies as the supporting technologies for macro-aspect methods. All the energy aware technologies are considered in the service point of view, so that energy aware issue can be solved more systematically. The contributions of this paper are as follows:Firstly, an energy consumption assessing model is proposed, which gives a method for the network to make a decision whether to use data compression or data aggregation technology. After that, an aggregation tree constructing algorithm based on energy consumption assessment is proposed, which can get the information of hops from the aggregation point to the sink and the hops from the aggregation point to the source node is used to construct such an aggregation tree. Meanwhile, an approach to reduce the cost of reinforcement is also presented, in which the reinforcement work is done by the source nodes themselves, not by the sink node. Simulation results show that this approach can save more energy than GIT when the aggregation ratio is small, and that the further the sources are away from the sink, the less reinforcement messages are needed.Secondly, a routing algorithm is proposed that can make better choices is proposed, which is modeled by dynamic game theory takes the available power of the nodes and the power consumption of the path into consideration. This method is more predictable for being combined with sleep scheduling scheme and thus can prolong the lifetime of WSN, which overcomes the random defect of Energy Aware Routing. The method is also easy to be combined with sleep scheduling technologies to prolong the lifetime of WSN.Thirdly, in microscopic energy-saving area, which takes energy-saving as its specific need, one way to reduce information content in localization technology is studied, taking network localization technology as an example. The problem of energy aware localization is a tradeoff between the amount of information and the accuracy of localization. The constraint unique realization is proposed to make the positions of the network nodes determined, which also decreases the upper bound of energy consumption burning on localization. The lower bound of connectivity is weakened by our theory, which means there will be fewer collisions and further means less energy consumption.Lastly, a lifetime optimizing scheme is proposed to combine different applications, network service, network deployment and optimizing algorithms together. The statistic model is employed to unify different applications and optimizing algorithms in this scheme, thus application requirements and network service ability are quantified. In every link of the network solutions, various energy aware technologies are considered in the service point of view, which lays a foundation for a more independent requirement of the solutions.This paper studies the three most representative energy-saving technologies in microscopic energy-saving area, aiming at getting a clear picture of the energy-saving issue in general. So in macroscopic energy-saving area, this paper focuses on service-oriented network lifetime, viewing the three technologies mentioned above as the supporting technology of the entire network energy-saving. Since various energy-saving technologies are taken into consideration from the perspective of service, the energy-saving of WSN is more systematized. This study of service-oriented optimizing scheme of network lifetime enables energy-saving to become the system design for the whole lifetime, which helps establish energy-saving as a relatively independent research area. Thus, energy-saving can be proposed as an independent system requirement like“network security”, consequently facilitating the systematization of solution.更多还原