移动社会网络拥塞控制研究

发布时间：2018-05-27 11:36

本文选题：移动社会网络 + 拥塞控制　；参考：《北京邮电大学》2014年硕士论文

【摘要】：随着社会网络服务的普及和移动终端技术的不断发展,一种新型网络开始出现。这种网络往往由人类携带智能通信设备,如手机,平板等,通过自组织的形式组成,同时具有移动性和社会性的特点,我们一般将这类网络称之为移动社会网络(Mobile Social Network,MSN)。由于人类的移动特性、设备能量受限等各种原因,节点间的连接通常呈现出间歇性连接、不稳定等延迟容忍网络(Delay Tolerant Network,DTN)的特征。 MSN通常利用节点社会度量辅助进行下一跳转发节点的选择,虽然可以提高数据的交付率,却带来了一定的负面影响,网络负担会较快的向活跃的节点集中,造成某些活跃节点缓存资源的耗尽,进而导致网络的拥塞。因此,研究适用于移动社会网络特点的拥塞控制机制具有重要的意义。本文将针对移动社会网络的特点,建立网络模型,并在此基础上,研究优化目标为源节点级公平性的拥塞控制机制,在提高网络吞吐量的同时保证网络不发生拥塞,最终提高网络的性能。本文研究内容和主要贡献包括： 1.分析移动社会网络中拥塞控制主要研究的问题,分析其研究方法与主要技术,总结研究现状。 2.针对目前拥塞控制研究问题的限制,结合社会性路由和多副本路由的特点,提出一种移动社会网络模型,为后续拥塞控制的研究提供基础。 3.结合所提网络模型,从源节点副本配额分配入手,将副本配额分配问题形式化为多维多选择背包问题,提出一种优化目标为源节点级公平性的拥塞控制机制。 4.根据理论分析进行算法设计和实现,在ONE仿真软件平台进行仿真验证并对结果进行分析,结果表明该机制可以较好的改善网络性能。
[Abstract]:With the popularity of social network services and the continuous development of mobile terminal technology, a new type of network began to emerge. This kind of network is usually composed of intelligent communication devices, such as mobile phone, flat panel, etc. It has the characteristics of mobility and sociality. We generally call this kind of network Mobile Social Network / MSN. Due to the mobility of human beings and the limited energy of devices, the connection between nodes usually presents the characteristics of intermittent connection and unstable delay tolerance network (Delay Tolerant Network). MSN usually uses node social metrics to assist in the selection of the next hop forwarding node. Although it can improve the data delivery rate, it brings some negative effects, and the network burden will quickly focus on the active nodes. This results in the depletion of cache resources of some active nodes, which leads to network congestion. Therefore, it is of great significance to study congestion control mechanisms suitable for mobile social networks. In this paper, according to the characteristics of mobile social networks, a network model is established, and on this basis, a congestion control mechanism which optimizes fairness at the source node level is studied to improve network throughput and ensure that network congestion does not occur. Finally, the performance of the network is improved. The contents and main contributions of this paper are as follows: 1. This paper analyzes the main research problems of congestion control in mobile social networks, analyzes its research methods and main technologies, and summarizes the present research situation. 2. In view of the limitations of congestion control research and the characteristics of social routing and multi-replica routing, a mobile social network model is proposed, which provides a basis for the further study of congestion control. 3. Based on the proposed network model, the replica quota allocation problem is formalized as a multi-dimensional multi-selection knapsack problem, and a congestion control mechanism is proposed to optimize fairness at source node level. 4. The algorithm is designed and implemented according to the theoretical analysis, and the simulation results are verified on the ONE simulation software platform. The results show that the mechanism can improve the network performance better.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP393.06

【参考文献】