当前位置:主页 > 科技论文 > 信息工程论文 >

空天地一体化网络路由协议的研究

发布时间:2018-02-14 06:17

  本文关键词: 空天地一体化 位置预测 DTN 数据束层 灰色预测 出处:《电子科技大学》2016年硕士论文 论文类型:学位论文


【摘要】:空天地一体化网络是一种融合天基、空基、地基网络,实现各层次间互连互通的网络应用场景。其有效的发展对军事、民用、空间科学探索具有极大的推动作用,所以空天地一体化网络的发展代表着一个国家的信息化的进步程度。为了应对各种环境条件下的通信局限性,不断涌现的针对某单一环境条件下的网络架构(受限网络),独立的协议架构为全网的互连、互通提出了挑战。不同的协议架构往往具有各异的传输层以及底层网络,并且编址方案也难以统一。因此如何实现各个独立网络域之间的互通和有效的路由成为了学者们研究的热点。本文结合具体的网络场景(空天地一体化网络),借鉴DTN的体系架构,设计了可以覆盖多种网络,并能够在各不同子网域之间传输数据的空天地一体化网络体系架构。在协议架构的传输层和应用层之间实现了数据束层,这种基于存储-转发的设计,最终实现了数据的跨域传输,并且屏蔽了底层网络的不同。由于空间通信具有频繁中断大延迟的物理特点,导致传统的基于通信过程中具有连通图的路由算法(OLSR,AODV等)无法实现数据的有效转发。本文提出CGR-EPI的分层路由方案,由数据束层的路由协议在不同网络层次选择合适的路由方式。对全网DTN节点利用URI方案统一编址,实现了目的地址的后绑定,解决了DNS在中断网络下无法正常运行的缺点。最后鉴于EPI路由算法的网络负载过大容易造成缓存不足等缺点,提出了基于灰色预测的多副本路由算法。利用灰色预测算法对节点运动平面坐标系中节点位置的横纵坐标的下一时刻值进行预测。经过对算法性能仿真和对应路由性能仿真,结果表明经过预测位置的路由算法有效地防止了由于节点密集时造成的路由风暴,提高了数据包到达率,降低了网络负载。
[Abstract]:The space-earth integrated network is a network application scene that integrates space-based, space-based and ground-based networks and realizes interconnection between different levels. Its effective development has a great impact on the exploration of military, civil and space science. Therefore, the development of the integrated network of space, earth and earth represents the degree of information progress of a country. In order to cope with the limitations of communication under various environmental conditions, The emerging network architecture for a single environment (limited network architecture, independent protocol architecture for the interconnection of the whole network, interworking has posed a challenge. Different protocol architectures often have different transport layer as well as the underlying network. Therefore, how to realize the interworking and effective routing among all independent network domains has become a hot research topic of scholars. This paper combines the specific network scenarios (integrated network of space, earth and earth). Looking at the architecture of DTN, An integrated network architecture, which can cover multiple networks and transmit data between different sub-domains, is designed, and the data bundles layer is implemented between the transport layer and the application layer of the protocol architecture. The design based on storage and forwarding finally realizes the cross-domain transmission of data and shields the differences of the underlying network. Because space communication has the physical characteristics of frequent interruptions and large delays. As a result, the traditional routing algorithm with connected graph in communication process, such as OLSR-AODV, can not effectively forward the data. This paper proposes a hierarchical routing scheme for CGR-EPI. The routing protocol of the data bundle layer selects the appropriate routing methods at different network levels. For the DTN nodes in the whole network, the URI scheme is used to uniformly address the destination address, and the post-binding of the destination address is realized. It resolves the shortcoming that DNS can not run normally under the interrupt network. Finally, considering that the network load of EPI routing algorithm is too large, it is easy to cause the shortage of cache, and so on. This paper presents a multi-replica routing algorithm based on grey prediction, which is used to predict the next moment value of the vertical coordinate of node position in the moving plane coordinate system of nodes. Through the simulation of the algorithm performance and the corresponding routing performance simulation, The results show that the routing algorithm can effectively prevent the routing storm caused by the dense nodes, improve the packet arrival rate and reduce the network load.
【学位授予单位】:电子科技大学
【学位级别】:硕士
【学位授予年份】:2016
【分类号】:TN915.04

【参考文献】

相关期刊论文 前2条

1 袁基炜;史忠科;;一种基于灰色预测模型GM(1,1)的运动车辆跟踪方法[J];控制与决策;2006年03期

2 金义富;朱庆生;邢永康;;序列缺失数据的灰插值推理方法[J];控制与决策;2006年02期



本文编号:1510067

资料下载
论文发表

本文链接:https://www.wllwen.com/kejilunwen/xinxigongchenglunwen/1510067.html


Copyright(c)文论论文网All Rights Reserved | 网站地图 |

版权申明:资料由用户8fb0c***提供,本站仅收录摘要或目录,作者需要删除请E-mail邮箱bigeng88@qq.com