容迟网络中多策略路由算法的设计与实现

发布时间：2018-05-28 13:42

本文选题：容迟网络 + 路由协议　；参考：《青岛大学》2017年硕士论文

【摘要】：容迟网络作为一种新型无线移动网络,其特点是节点移动造成缺乏持久的端到端连接、动态变化的网络拓扑结构、有限的存储空间等。为了解决链路缺失等问题,实现容迟网络中不同网络构架之间的互联和互操作,容迟网络采用“存储-携带-转发”的消息传输机制。为此,如何做出高效准确的路由决策是实现信息共享及高效传输的关键技术。本论文研究如何设计适用于容迟网络的高效路由算法,主要内容有以下三点:(1)基于节点相似性的概率路由算法。通过研究节点的历史相遇信息,挖掘节点社交范围的相似性。利用统计学方法计算节点间的接触频率,在此基础上预估节点在消息生存时间内与目的节点的相遇概率,实现消息定向扩散。(2)基于节点社交流行性路由。通过记录节点的历史移动轨迹,获取与给定节点的紧密连接邻居节点。克服度中心性只考虑一跳连接的缺点,从而提出启发性函数节点流行性。(3)基于节点社交流行性的概率路由算法。提出节点社交群感知模型,并在此基础上定义启发式函数。综合考虑消息传输时延及消息剩余生存周期,预估节点与目的节点直接与间接的相遇概率。以消息投递率为优化目标,把中继节点选择问题转化为0-1背包的最优化问题,通过回溯算法求解。该算法将路由策略分为基于目的节点社交群内路由和基于目的节点社交群外路由两个阶段。通过大量仿真实验,验证了论文工作的高效性及合理性。提出的三种路由算法分别在消息投递率、网络负载等方面取得较好的性能表现,对容迟网络中路由算法研究领域具有较高的理论价值及参考价值。
[Abstract]:As a new type of wireless mobile network, delay-tolerant network is characterized by the lack of end-to-end connection caused by node mobility, dynamic network topology, limited storage space and so on. In order to solve the problem of missing links and realize the interconnection and interoperation among different network architectures in delay-tolerant networks, the delay-tolerant networks adopt the mechanism of "store, carry and forward" message transmission. Therefore, how to make efficient and accurate routing decision is the key technology to realize information sharing and efficient transmission. This paper studies how to design an efficient routing algorithm for delay tolerant networks. The main contents are as follows: 1) probabilistic routing algorithm based on node similarity. By studying the historical encounter information of nodes, the similarity of social scope of nodes is mined. The frequency of contact between nodes is calculated by statistical method, and then the probability of the nodes meeting with the destination node within the message lifetime is estimated, and the message oriented diffusion. 2) based on the node social epidemic routing is realized. By recording the historical movement path of the node, the tightly connected neighbor node with a given node is obtained. To overcome the shortcoming of degree centrality considering only one-hop connection, a probabilistic routing algorithm based on node social popularity is proposed for heuristic function node epidemic. A nodal social group perception model is proposed and a heuristic function is defined on this basis. The probability of direct and indirect encounter between the node and the destination node is estimated by considering the residual life cycle of the message when the message is transmitted. With message delivery rate as the optimization objective, the relay node selection problem is transformed into the 0-1 knapsack optimization problem, which is solved by backtracking algorithm. The algorithm divides the routing strategy into two stages: the intra-group routing based on the destination node and the out-of-group routing based on the destination node. Through a large number of simulation experiments, the high efficiency and rationality of the paper work are verified. The proposed three routing algorithms have achieved better performance in message delivery rate and network load respectively, which is of high theoretical value and reference value to the research field of routing algorithms in delay-tolerant networks.
【学位授予单位】：青岛大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP393.02

【参考文献】