【作者】 张可；

【导师】 曾家智；

【作者基本信息】 电子科技大学 ， 计算机系统结构， 2010， 博士

【摘要】 随着现代科学技术的不断发展,人们对网络通信技术的需求也不断提升,新的应用不断涌现,传统网络已经不能满足社会的需求,从而促进了无线网络技术的快速发展。无线自组织网络由于其具有自组织、自愈合、快速反应等特性,在各行各业中的应用日益广泛。特别是近来人们关注较多的车载移动自组织网络、无人机载自组织网络等,有重大的理论和现实意义。无线传感器网络作为一种特殊的自组织网络,已经在军事、交通监控、安全敏感区域监控、智能家居和目标跟踪、公共安全检测等领域有广泛的应用前景。因此无线移动组织网络及传感器网络成为了无线自组织网络技术研究中的热点问题,由于其特殊性,因此具有一定的研究挑战。本文在对无线自组织及传感器网络进行了系统全面的分析和总结基础上,对研究热点问题中的延迟容忍移动传感器网络数据传输策略、有向无线传感器网络覆盖问题及移动自组织网络路由等几个方面的内容深入细致地开展了相关研究工作,并取得了若干的创新与成果。本文的主要贡献包括:1.针对延迟容忍移动传感器网络的特点,对其现有的数据传输策略进行了深入的研究,并在经典延迟容忍移动传感器网络数据传输策略FAD以及MPAD策略的基础上先后提出了一种基于优先度复制的数据传输策略PRD用于空间中间断连通的延迟容忍移动传感器网络数据传输以及一种基于数据融合的适配数据复制传输策略DAAD,该传输策略主要用于采集监控环境数据的延迟容忍移动传感器网络。PRD由选择复制策略和队列管理组成,前者根据节点将消息传递给汇聚点的可能性,选择下一跳进行复制传输;队列管理则利用引入传输优先度及复制数的消息生存时间决定队列中消息丢弃原则。DAAD由选择复制策略和具有数据融合功能的队列管理组成,前者适应性的选择下一跳传输的节点,复制消息进行数据传输;队列管理则决定队列中消息的生存与否等,同时利用数据融合,产生新消息,减少网络中多余数据消息的传输。仿真分析表明,PRD以及DAAD均能够有效提高数据交付率,同时具有较低的数据复制数以及传输延迟。2.针对有向传感器网络覆盖增强问题,提出了基于人工鱼群优化的有向传感网络覆盖增强算法AFCEA以及基于遗传模拟退火算法的有向传感器网络覆盖增强算法GSACEA。在AFCEA中,将针对性改进的人工鱼群优化引入可转动模型下有向传感器网络覆盖增强问题最优解的求解,通过多条人工鱼的一系列行为,逐步接近网络覆盖的最大值。通过理论研究以及仿真实验分析,与其他典型的有向传感器网络覆盖增强算法相比,AFCEA算法可以以较少的迭代计算次数有效提高有向传感器网络的监测覆盖率。GSACEA算法则将遗传算法与模拟退火算法相结合并作了一定的改变,通过遗传的进化作用与模拟退火的作用逐步接近最优解,解决有向传感器网络覆盖增强问题。仿真验证表明,与其他经典的同类研究算法相比,GSACEA可以以较少的迭代次数有效提高有向传感器网络监测覆盖率。3.在人工鱼群算法的基础上,创新性地提出了虚拟水流的概念,分析了虚拟水流对人工鱼群算法的影响,进而提出了一种基于虚拟水流人工鱼群优化的有向传感网络覆盖增强算法VSAFCEA。在VSAFCEA中,将引入虚拟水流并重新定义行为模式的人工鱼群优化引入可转动模型下有向传感器网络覆盖增强问题最优解的求解,通过多条人工鱼在虚拟水流下的一系列行为,逐步接近网络覆盖的最大值。通过理论研究以及仿真实验分析,与其他有向传感器网络覆盖增强算法,包括AFCEA相比,VSAFCEA算法可以以较少的迭代计算次数有效提高有向传感器网络的监测覆盖率。4.对无人机自组织网络等高速移动环境下的无线移动自组织网络特性进行了相关分析,针对节点移动快速、网络拓扑变化剧烈等特性,提出了一种上下文感知优化链路状态协议CAOLSR,采用了一种上下文信息机制,引入了节点地理信息位置等信息,并综合多个网络上下文影响参数,将节点间相对移动预测、前后访问时间以及节点连接度情况引入MPR选择,并设计了CAOLSR-MPR算法流程。此外,CAOLSR还通过引入Fisheye技术,分层次对传输控制消息的发送频率进行了控制,减少了移动性对路由精度的影响。最后通过多组仿真实验表明,在节点快速移动与拓扑快速变化环境下与HOLSR、OLSR、DSDV相比,CAOLSR具有更为良好的性能。在本文的最后,总结了全文的工作,并对未来的研究工作进行了展望。更多还原

【Abstract】 With the development of modern network technology, the requirements of human society for communication technology in usual life and new applications are proposed unceasingly. Traditionary networks can’t fit the requirements of modern society, so the wireless networks are developed rapidly.Wireless Ad Hoc networks are applicated in our society more and more, because they have some characteristics, such as self-organization, self-concrescence and rapid reaction. Wireless Ad Hoc networks have very important significance in theory research and application, especially in some domains these we put more attentions, such as vehicular mobile Ad Hoc networks and Unmanned Aerial Vehicles networks. As a special Ad Hoc network, wireless sensor networks have many applications in military affairs, traffic, special area stakeout, intelligence family, target tracking and public security. So wireless mobile Ad Hoc networks and sensor networks are two hot research domains in wireless Ad Hoc networks, for these special characteristics, there are some research challenges.Based on a systematical summary of wireless Ad Hoc networks and sensor networks, this dissertation focuses on some related sub-researches, such as data delivery scheme of delay tolerant mobile sensor networks, coverage-enhancing of directed sensor networks and routing of mobile Ad Hoc networks , and gains several innovations and achievements. Our research works and contributions of this dissertation are summarized as follows:1. In this dissertation, we propose a new data delivery scheme-PRD (Priority Replication Delivery scheme) for pervasive data gathering in DTMSN (Delay Tolerant Mobile Sensor Networks) that network with intermittent connectivity in space, and we propose another scheme-DAAD(Data Aggregation-based Adaptive Data Reproductive Delivery Scheme) for environment monitoring in DTMSN. PRD consists of two key components for data transmission and queue management, respectively. The former makes decisions on when and where to transmit data messages according to the node delivery probability. The latter employs the message survival time based on priority and delivery copies to decide dropping for minimizing transmission overhead. DAAD consists of data transmission scheme and queue management based on data fusion, respectively. The former makes decisions on when and where to transmit data messages. The latter manages the message queue based on data. Simulation results show that the proposed two data delivery schemes achieve the higher message delivery ratio with the lower transmission overhead and data delivery delay than other DTMSN data delivering approaches.2. In this dissertation, we propose an AFCEA (Artificial Fish-swarm based Coverage-Enhancing Algorithm) and a GSACEA (Genetic Simalated Annealing based Coverage-Enhancing Algorithm) for the problem of coverage-enhangcing in directed sensor networks. Directed sensor is a kind of directional system, and its coverage is different with omni-directional system. In AFCEA, we use our improved optimization algorithm into the solution for coverage-enhancing in directed sensor networks with rotational direction model. By analysis and some simulations, and compare AFCEA to other classic directed sensor networks coverage-enhancing algorithm, AFCEA can achieve higher directional sensor networks coverage with lesser iterative computing times. GSACEA combine the genetic algorithm and the simulated annealing algorithm into the solution for coverage-enhancing in directed sensor networks. By simulations, and compare to other classic directed sensor networks coverage-enhancing algorithm, GSACEA can achieve higher directional sensor networks coverage with lesser iterative computing times too.3. Based on artificial fish-swarm algorithm, in this dissertation we propose a concept named virtual stream, and analyse virtual stream’s influence on artificial fish-swarm algorithm, then we propose a VSAFCEA (Virtual Stream Artificial Fish-swarm based Coverage-Enhancing Algorithm). In VSAFCEA, we use our improved artificial fish-swarm optimization algorithm with virtual stream into the solution for coverage-enhancing in directed sensor networks with rotational direction model. By analysis and some simulations, and compare VSAFCEA to other classic directed sensor networks coverage-enhancing algorithm and AFCEA, VSAFCEA can achieve higher directional sensor networks coverage with lesser iterative computing times.4. In this dissertation, a Context-aware optimized link state routing protocol for networks with fast-moving nodes is proposed; CAOLSR (Context-aware Optimized Link State Routing Protocol) adopts a context-aware mechanism, and selects MPR (Multi Point Relays) based on relative movement of nodes, recent access-time and connection number of nodes, and adopts a special flow of MPR selection. In addition, by the introduction of Fisheye, reduces the influence from mobility on the routing accuracy. Experimental results have shown that CAOLSR can achieve good performance and outperform HOLSR (Hierarchical Optimized Link State Routing Protocol), OLSR (Optimized Link State Routing Protocol) and DSDV (Destination Sequenced Distance Vector) in networks with fast-moving nodes.Summaries and prospects have also been put forward in the final of this dissertation.更多还原