基于智能优化算法的空基网路由算法研究

发布时间：2018-04-20 08:00

本文选题：空基网 + 路由算法　；参考：《哈尔滨工业大学》2017年硕士论文

【摘要】：随着航空航天和空间探索领域的快速发展,空间信息的影响力变得越来越高,对空间信息的掌握程度,从某种程度上会影响一个国家综合国力的提升和社会经济的发展。空天地一体化网络由深空网络、近地空间网络、地面网络构成,其中近地空间部分由飞机、直升机、战斗机、无人机等低空飞行器构成,也被称为空基网。空基网中的各种飞行器节点都具有实时移动性,且节点运动速度很高,导致网络拓扑结构频繁变化,因此,空基网的路由技术面临着很大的问题,包括路由有效时间短、通信路由频繁断开、数据包传输时延长等。要采取适应空基网特征的路由算法,要求路由算法收敛速度快,路由建立时间短,平均路由跳数小,路径优化速度快。本文主要研究空基网飞行器之间基于智能优化算法的路由算法。首先,针对空基网节点移动性高、拓扑变化频繁的特点,将蚁群算法、粒子群算法、遗传算法应用到网络中,建立智能优化路由算法。将三种智能优化路由算法与AODV协议仿真对比,结果表明蚁群优化路由算法在路由建立成功率、平均路由跳数性能上表现最好,粒子群优化路由算法和遗传优化路由算法次之,而AODV协议最差。接着,考虑网络的簇头节点不再位于网络中心位置或者失效、毁坏、突然退出网络的场景,设计基于图论的簇头节点选择和更新算法,并使用概率图模型知识对其进行改进,以得到更加符合空基网场景的结果,提出基于概率图模型的簇头节点选择和更新算法。分析表明基于概率图模型的簇头节点选择和更新算法尽管比基于图论的方法算法复杂度略高,但是其簇头与网络内节点的平均路由跳数更少。最后,对有新节点加入网络的情形,利用概率图模型理论,综合考虑迟入节点与其邻接点的相互关系,为迟入节点选择最佳的接入节点。仿真结果表明,该方法选择的接入节点与按照距离最近原则选出的接入节点有所不同。
[Abstract]:With the rapid development of aerospace and space exploration, the influence of space information becomes more and more high. To some extent, the mastery of space information will affect the promotion of a country's comprehensive national strength and the development of social economy. The integrated network is composed of deep space network, near-earth space network and ground network, in which the near-Earth space is composed of aircraft, helicopters, fighter planes, drones and other low-altitude aircraft, also known as space-based network. All kinds of aircraft nodes in space-based networks have real-time mobility, and the speed of node movement is very high, which leads to frequent changes in network topology. Therefore, the routing technology of space-based networks faces great problems, including the short effective time of routing. The communication route frequently disconnects, the packet transmission time lengthens and so on. In order to adopt the routing algorithm which adapts to the characteristics of space-based network, it is necessary for the routing algorithm to converge fast, to set up the route in a short time, to reduce the average number of hops, and to optimize the route quickly. In this paper, the routing algorithm based on intelligent optimization algorithm is studied. Firstly, the ant colony algorithm, particle swarm optimization algorithm and genetic algorithm are applied to the network to establish an intelligent optimal routing algorithm. The simulation results show that the ant colony optimization routing algorithm has the best performance on the average number of hops, the particle swarm optimization routing algorithm and the genetic optimization routing algorithm are the second, compared with the simulation of AODV protocol, the ant colony optimization routing algorithm has the best performance in routing establishment, and the performance of the average number of hops is the best. AODV protocol is the worst. Then, considering the cluster head node is no longer located in the center of the network or failure, destroy, suddenly quit the network scene, design a graph based cluster head node selection and update algorithm, and use the probability graph model knowledge to improve it. Based on probability graph model, a cluster head node selection and update algorithm is proposed in order to obtain the results that are more consistent with the space-based network scenario. The analysis shows that the clustering head selection and updating algorithm based on probabilistic graph model is a little more complex than the one based on graph theory, but the average number of routing hops between cluster heads and nodes in the network is less. Finally, in the case of adding new nodes to the network, the theory of probabilistic graph model is used to comprehensively consider the relationship between late entry nodes and their adjacent points to select the best access nodes for late entry nodes. The simulation results show that the access nodes selected by this method are different from those selected according to the principle of distance nearest.
【学位授予单位】：哈尔滨工业大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：V443.1;TP18;TN915.0

【参考文献】