信息中心网络中认知型路由协议的设计与仿真实现

发布时间：2018-11-14 12:32

【摘要】：近年来,为了解决以地址为中心的网络通信模型通信效率不高的问题,学术界提出了一种以信息为中心的新型网络体系架构一信息中心网络(Information-Centric Networking, ICN),信息中心网络使用以内容为核心的协议栈结构来替换传统的以IP为核心的协议栈结构。针对这种新型网络体系结构,如何设计一种能够充分发挥网络自身优势的路由协议是信息中心网络研究中亟待解决的关键问题。本文在信息中心网络节点加入了认知智能和社交能力,提出了一种认知型路由协议。首先,针对信息中心网络的特点,本文设计了网络的路由器结构和网络模型；其次,设计了网络节点的基本结构,包括缓存(ContentStore, CS)、未决兴趣表(Pending Interest Table, PIT)和转发信息表(Forwarding Information Base, FIB);最后,在上述基础之上,将感知、询问、学习、推理和反馈5种认知行为引入到信息中心网络的路由协议中,并使网络中的节点拥有社交能力。通过感知行为,节点感知到邻居和相邻链路状态、本地网络拓扑信息和流行内容；通过询问行为,节点询问得到邻居缓存表和朋友缓存表；通过学习行为,节点学习到请求经验路径信息和响应经验路径信息；通过推理行为,节点推理出粉丝关系和朋友关系,进而获取关注内容列表和朋友缓存列表；通过反馈行为,节点利用以上认知行为获取的认知信息做出路由决策。网络中的节点拥有社交能力,节点在路由过程中可以询问自身社交圈中的邻居节点、关注源节点和朋友节点,获取对路由转发有帮助的信息。本文对提出的信息中心网络中认知型路由协议进行了仿真实现和性能评价。其结果表明,本文提出的认知路由协议能充分利用信息中心网络的特点,并很好地将认知智能和社交能力融入到信息中心网络中,同时协议在启动阶段、节点加入或退出阶段以及协议稳定阶段都有比较好的性能,是有效可行的。
[Abstract]:In recent years, in order to solve the problem of low communication efficiency in the address-centric network communication model, the academic community has proposed a new information-centric network architecture, the Information Center Network (Information-Centric Networking, ICN),). The information center network uses the content-centered protocol stack structure to replace the traditional protocol stack structure with IP as the core. In view of this new network architecture, how to design a routing protocol which can give full play to the advantages of the network itself is the key problem to be solved urgently in the research of information center network. In this paper, cognitive intelligence and social ability are added to the information center network node, and a cognitive routing protocol is proposed. Firstly, according to the characteristics of the information center network, this paper designs the router structure and the network model of the network. Secondly, the basic structure of network nodes is designed, including caching (ContentStore, CS), pending interest table (Pending Interest Table, PIT) and forwarding information table (Forwarding Information Base, FIB);). Finally, the cognitive behaviors of perception, inquiry, learning, reasoning and feedback are introduced into the routing protocols of the information center network, and the nodes in the network have social ability. Through sensing behavior, nodes perceive neighbor and adjacent link state, local network topology information and popular content, by questioning behavior, the node interrogates to obtain neighbor cache table and friend cache table. Through the learning behavior, the node learns the information of the request experience path and the response experience path information, through the reasoning behavior, the node deduces the fan relationship and the friend relation, and then obtains the list of attention content and the friend cache. Through feedback behavior, nodes make routing decisions using the cognitive information obtained from the above cognitive behaviors. The nodes in the network have the social ability. In the process of routing, the nodes can ask their neighbors in the social circle, pay attention to the source node and the friend node, and obtain the information that is helpful to route forwarding. In this paper, the proposed cognitive routing protocol in the information center network is simulated and evaluated. The results show that the proposed cognitive routing protocol can make full use of the characteristics of the information center network, and integrate cognitive intelligence and social ability into the information center network, and at the same time, the protocol is in the start-up stage. The node join or exit phase and the protocol stabilization phase have better performance and are effective and feasible.
【学位授予单位】：东北大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TN915.04

【相似文献】