【作者】 鲍彦茹；

【导师】 舒炎泰；

【作者基本信息】 天津大学 ， 计算机应用技术， 2008， 博士

【摘要】 近几年,无线Ad Hoc网络得到了迅速的发展,并已经被广泛地应用于军事和救援等应急通信中。路由选择问题,是Ad Hoc网络的一个重要和核心问题。同时,由于Ad Hoc网络中移动终端均由电池供电,而电池容量有限,因此,Ad Hoc网络中节能路由协议的设计和改进具有非常重要的研究意义和应用价值。大部分的Ad Hoc网络数据链路层协议都提供了保证数据可靠传输的重传机制。重传机制势必引起能量花费的增加,而现有的能量有效路由算法在计算能量花费时大都省略了这部分花费。因此,本文提出了一种基于AODV的节能路由算法REAODV。在选择路由时,该算法将不可靠链路上进行可靠数据传输所消耗的总能量(包括重传所消耗的能量)作为选路指标。大量的仿真表明,该算法能有效的节省节点能量,延长了网络的生存时间。无线Ad Hoc网络节能路由协议一般以两个目标进行设计,即最小化能量花费和最大化网络生命期。现有节能路由协议仅单纯满足一个设计目标。因此,本文提出了基于表驱动的节能路由协议REARP。在寻路过程中,该路由协议综合考虑了最大化网络生命期和最小化每个包的能量花费。同时,该协议考虑了链路的可靠性,采用了功率控制技术,减少了包重传的次数。大量仿真表明,与现有路由协议相比较,该协议在保证网络性能的前提下,能够有效的节省能量,从而延长了网络生命期。针对Ad Hoc网络中存在多个路由请求的情况,本文讨论了最大化网络生命期的路由选择问题,并证明了一般情况下的最大生命期问题是NP完全问题,还证明了几何的最大生命期路由问题也是NP完全问题。对于该路由问题,本文提出了最大生命期路由算法DNLBR和IDNLBR。DNLBR算法以节点的生命期作为选路标准,寻找源节点与目的节点间路径生命期最大的路由。为了克服DNLBR算法延迟大的缺点,本文又提出了IDNLBR算法。通过迭代方式,该算法不断使用生命期较大的节点替换路由上生命期较小的节点,并通过限制跳数来减少路由延迟。仿真结果表明,上述两种算法优于现有路由算法,在电量一定的情况下,能够成功传输更多的数据包。评价和比较Ad Hoc网络中各种能量有效协议节能效果的关键在于建立能够准确计算能量花费的仿真模型。本文总结了现有的各种能量模型,分析了各模型的优缺点,并提出了线性能量模型的改进方案。该方案不仅能够计算出所有通信状态(包括发送、接收、空闲和睡眠)的能量花费,而且对于发送状态,能够计算出不同发射功率下的能量花费。更多还原

【Abstract】 Wireless ad hoc networks are useful in situations where temporary communication is needed, such as in law enforcement or in disaster relief. So the ad hoc networks draw lots of attention in recent years. Most of the studies performed in the field of wireless ad hoc networks have focused on the problem of routing. Since most ad hoc mobile devices today operate on batteries, energy efficient routings have drawn considerable research interests recently.In fact, the wireless communications are unreliable and often unpredictable, and almost all of current data link layer protocols support the retransmission mechanism. However, existing energy-aware routing protocols ignore the energy consumption on the retransmission. In this thesis, we propose an AODV based energy-efficient routing algorithm REAODV. The main characteristic of REAODV is that it uses the total energy consumption, including the retransmission energy cost, as the metric when choosing routings. Extensive simulations are conducted to show that our new routing algorithm can indeed save more energy than original routing protocols in ad hoc networks.In many scenarios, design of energy-efficient protocols is guided by two essential requirements: minimizing the overall transmission power for each connection request and maximizing the lifetime of ad hoc mobile networks. We present a new energy-aware routing algorithm called Reliable Energy-Aware Routing Protocol (REARP) which can satisfy these two objectives simultaneously. Moreover, dynamic power control, which is realized by cross-layer optimization, is used in the algorithm to reduce the energy cost. Simulation results show that network lifetime of REARP is extended for random traffics while its total energy consumption is decreased.We are also focused on the routing problems for multiple requests with the aim of maximizing network lifetime. We show that this maximum lifetime routing problem is NP-complete, both for the general case and for the geometric one. We then develop a new maximum lifetime routing algorithm called DNLBR, which directly uses the node lifetime as cost metric. Moreover, we improve the algorithm to IDNLBR, which overcomes the large delay of DNLBR by integrating minimum-hop algorithm. To found the maximum lifetime routes for multiple requests networks, IDNLBR iteratively modifies the route by substituting long lifetime nodes for short lifetime nodes. By way of simulations, we compared our algorithms to existing protocols, and the results show that our algorithms are better at extending network lifetime. In order to evaluate and compare the energy-aware protocols in terms of their energy efficiency, an energy-consumption model which can accurately computes the energy consumed by the data communication activities is crucial. We give a comprehensive summary of the existing energy-consumption models, and the characteristics of each model are discussed in detail. We propose a new energy-consumption model which is based on the linear model. All nodes in the new model can change their RF power-rate and radio states according to the communication requirement, and their energy cost can be calculated correctly.更多还原