基于区域朋友关系的机会网络路由算法的研究

发布时间：2018-05-14 01:29

本文选题：机会网络 + 亲密程度　；参考：《南京邮电大学》2017年硕士论文

【摘要】：近年来,移动通信已经逐渐成为一种人们在移动中随时共享信息和获取服务的重要方式。传统无线通信方式是借助一定的基础设施(如基站)来保障通信节点之间的连通性来完成通信的,但是在一些无法部署基础设施的恶劣环境,传统无线通信方式就会失效。为了解决此问题,一种新型的网络移动机会网络(Mobile Opportunistic Networks,MON)应运而生。MON中节点位置和网络规模频繁变化,端到端通信不需要确定是否存在通路,而是利用节点移动创造的相遇机会完成通信,因此它能够解决间断网络的通信难题,具有广阔的应用前景。但是拓扑结构的复杂多变、路由信息的缺乏等问题给MON的数据转发带来技术上的挑战,而机会路由算法可以通过对节点移动规律进行预测,使数据最大限度地转发至目的节点。一般地,对节点历史轨迹信息进行分析可以在一定程度上帮助预测节点的移动规律,此外由人携带的设备组成的MON还可以利用节点移动的社会属性来预测节点移动规律从而达到提高预测精度的目的。本文主要研究如何利用节点移动的社会属性来提高机会路由算法性能。基于社会属性的机会路由算法通常考虑的社会属性包括节点所属社区、节点亲密性、节点社会地位等。其中在利用亲密性预测节点接触概率的机会路由算法中,节点亲密性评价的准确性会影响节点接触概率预测的精度,进而影响机会路由算法的性能,因此应该全面分析与节点亲密性相关的各个因素,建立数学模型描述并整合各因素,期望能够准确地评价节点之间的亲密程度并设计相应的机会转发策略。现有的研究在评价节点亲密性时一般没有考虑节点位置影响。实际上,节点在不同区域其亲密节点也会不同。本文综合考虑了节点亲密性的时间相关性和区域相关性,利用节点的历史接触信息来构造节点亲密程度的评价模型,从而得到节点在相应区域的亲密节点作为其区域朋友,最后结合节点当前位置和区域朋友关系提出一种基于区域朋友关系的机会路由算法(Opportunistic Routing Algorithm Based on Regional Friendship,ORRF)。实验数据表明,在不同的节点分布密度和可容忍时延下,该算法均得到较高的数据送达率和较低的转发能耗。在本文的最后,设计实现了一个基于Wi-Fi P2P技术的可以进行点对点通信的机会网络Android原型系统,该系统具有周围节点搜索、与周围节点建立连接并进行数据传输的功能。最终运行演示机会网络原型系统,收集相关性能数据,为未来机会网络的研究提供数据支持。
[Abstract]:In recent years, mobile communication has become an important way for people to share information and access services at any time. The traditional wireless communication mode is based on certain infrastructure (such as base station) to ensure the connectivity between communication nodes to complete the communication, but in some bad environment where infrastructure can not be deployed, the traditional wireless communication mode will fail. In order to solve this problem, a new mobile opportunity network, Mobile Opportunistic Networks (mon), emerges as the times require. The node location and network size change frequently in mon, and end-to-end communication does not need to determine whether there is a path or not. Instead, it makes use of the encounter opportunity created by node movement to complete communication, so it can solve the communication problem of discontinuous network and has a broad application prospect. However, the complexity of topology and the lack of routing information bring technical challenges to the data forwarding of MON, and the opportunistic routing algorithm can make the data forward to the destination node to the maximum extent by predicting the law of node movement. Generally speaking, analyzing the historical track information of nodes can help to predict the movement law of nodes to some extent. In addition, the MON, which is composed of human devices, can also use the social attributes of node movement to predict the law of node movement so as to achieve the purpose of improving the prediction accuracy. This paper mainly studies how to improve the performance of opportunistic routing algorithm by using the social attributes of node movement. Opportunistic routing algorithms based on social attributes usually consider the social attributes including the community to which the nodes belong, the intimacy of the nodes, the social status of the nodes, and so on. In the opportunistic routing algorithm which uses closeness to predict node contact probability, the accuracy of node closeness evaluation will affect the accuracy of node contact probability prediction, and then affect the performance of opportunistic routing algorithm. Therefore, all factors related to node closeness should be comprehensively analyzed, and mathematical models should be established to describe and integrate the factors, in order to accurately evaluate the degree of closeness between nodes and design corresponding opportunity forwarding strategies. Existing studies generally do not consider node location when evaluating node closeness. In fact, nodes in different regions have different intimate nodes. In this paper, the temporal and regional correlation of node closeness is considered synthetically, and the evaluation model of node closeness is constructed by using the historical contact information of nodes, and the intimate nodes in the corresponding region are obtained as their regional friends. Finally, an opportunity routing algorithm based on regional friendships is proposed, which combines the current location of nodes and regional friendships. Experimental data show that the proposed algorithm can achieve higher data delivery rate and lower forwarding energy consumption under different node distribution density and tolerance delay. At the end of this paper, an opportunistic network Android prototype system based on Wi-Fi P2P technology is designed and implemented. The system has the function of searching the surrounding nodes, establishing connections with the surrounding nodes and transmitting data. Finally, the prototype system of demo opportunity network is run, and relevant performance data are collected to provide data support for future research of opportunity network.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【参考文献】