软件定义物联网路由研究

发布时间：2019-04-12 19:40

【摘要】：物联网和软件定义网络是新一代信息与网络的核心技术,尤其是计算机网络是当前的热门研究课题。在物联网时代,异构网络系统的物理基础设施将更加复杂,在物联网中引入软件定义网络,可以为物联网的感知层、网络层以及与边缘设备或传感器节点连接的核心蜂窝网络提供弹性、敏捷和个性化的管理机制。在智能应用和服务方面,如智慧城市和医疗监控,物联网将成为软件定义网络发展的主要推动力。从物联网架构的角度来看,无线传感网是物联网的基础,路由是无线传感器网络中一个重要的研究领域。目前软件定义无线传感网的研究方兴未艾,软件定义路由替代传统路由的方法正在研究中,部分传统路由算法实际也应用了软件定义的思想。为了更好地推进物联网路由的发展,本文将软件定义网络技术引入到物联网路由研究中。探讨了软件定义物联网的体系架构及架构流程,提出了一个软件定义无线多跳网络的路由协议,通过OPNET对所提出的协议进行了仿真验证,结果证明,利用SDN技术进行集中式路径规划较传统的分布式规划更利于实现全局最优化,降低了网络时延。协议可为节点提供到目标节点的单条最短路径或独立多径路由,在负载足够大时,网络寿命比其他算法(AODV,OLSR)更长。在物联网协作通信和智能代理方面,提出了一种新的协作通信算法,提出了可靠性预测模型,每跳选择k个协作节点协助转发数据,使闲置的节点进入休眠以节省能耗,能够在高动态的无线传感网保持较高的可靠性；研究了多智能代理并行访问多个源节点的路线规划问题。提出了定向源分组多代理算法,每个源节点的迭代算法限于一个扇区内部,路线的长度由定向扇区的角度控制,选择适当的角度可得到最优的路线。在物联网多媒体独立多径路由方面,根据B样条曲线拟合了多径独立路由,提出了多径规划路由算法,在保证用户时延需求的前提下使得能耗最小。应用软件定义网络思想,研究了多源多径的路由协调算法,通过Sink节点编程实现动态控制源节点的路径；针对汇聚节点周围容易形成能量空洞的问题,提出了紧耦合地理多径路由算法PWDGR,使汇聚节点附近邻居负载均衡,相比DGR显著延长了网络寿命。在软件定义物联网的仿真方法方面,提出了一种新的半实物的方法来仿真软件定义物联网路由,使用OPNET的SITL技术搭建了软件定义健康网络半实物仿真平台,传感网部分或全部使用软件仿真,传输网络和云数据中心使用实际的网络。在保证应用QoS需求的情况下充分节约了能耗。
[Abstract]:Internet of things and software defined network are the core technologies of the new generation of information and network, especially the computer network is a hot research topic at present. In the era of the Internet of things, the physical infrastructure of heterogeneous network systems will be more complex, the introduction of software-defined networks in the Internet of things, can be the perception layer of the Internet of things, The network layer and core cellular networks connected to edge devices or sensor nodes provide resilient, agile, and personalized management mechanisms. In smart applications and services, such as smart cities and medical monitoring, the Internet of things will be the main driving force behind the development of software-defined networks. From the point of view of Internet of things architecture, wireless sensor network is the foundation of Internet of things, and routing is an important research field in wireless sensor network. At present, the research of software-defined wireless sensor networks is in the ascendant. The method of software-defined routing instead of traditional routing is being studied, and some traditional routing algorithms have also applied the idea of software definition. In order to promote the development of Internet of things routing, software-defined network technology is introduced into the research of Internet of things routing. This paper discusses the architecture and flow of software-defined Internet of things (IoT), presents a routing protocol for software-defined wireless multihop networks, and validates the proposed protocol by OPNET. Compared with traditional distributed planning, centralized path planning based on SDN technology is more conducive to global optimization and reduces network delay. The protocol can provide a single shortest path or independent multipath route to the target node. When the load is large enough, the network life is longer than other algorithms (AODV,OLSR). In the aspect of cooperative communication and intelligent agent of Internet of things, a new cooperative communication algorithm is proposed, and a reliability prediction model is proposed, in which k cooperative nodes are selected for each hop to help forward data, so that idle nodes can go to sleep to save energy consumption. It can maintain high reliability in high dynamic wireless sensor network; The route planning problem of multi-agent accessing multiple source nodes in parallel is studied. A directional source packet multi-agent algorithm is proposed. The iterative algorithm of each source node is limited to the interior of a sector. The length of the route is controlled by the angle of the directed sector. The optimal route can be obtained by selecting the appropriate angle. In the aspect of Internet of things multimedia independent multi-path routing, multi-path independent routing is fitted according to B-spline curve, and multi-path planning routing algorithm is proposed, which minimizes energy consumption under the premise of guaranteeing user delay requirement. Based on the idea of software definition network, the routing coordination algorithm of multi-source and multi-path is studied, and the path of source node is dynamically controlled by Sink node programming. In order to solve the problem that energy holes are easily formed around the convergence nodes, a tightly coupled geographic multipath routing algorithm, PWDGR, is proposed to balance the load of neighbors near the convergence nodes, which significantly prolongs the network life compared with DGR. In the aspect of software-defined Internet of things simulation method, a new hardware-in-the-loop method is proposed to simulate software-defined Internet of things routing, and a software-defined healthy network hardware-in-the-loop simulation platform is built by using SITL technology of OPNET. Sensor networks use part or all of the software simulation, transmission networks and cloud data center using the actual network. In the case of ensuring the application of QoS requirements, the energy consumption is fully saved.
【学位授予单位】：华中科技大学
【学位级别】：博士
【学位授予年份】：2016
【分类号】：TP391.44;TN929.5

【相似文献】