接触可预测的认知自组织网络动态多路径机会路由模型及其稳定性度量

发布时间：2018-11-06 15:38

【摘要】：随着无线通信技术的迅猛发展,可用的频谱资源变得越来越紧张,尤其是随着无线局域网、无线个域网和无线城域网等技术的高速发展,目前的宽带无线应用技术已无法满足人们的需求。目前的频谱分配体制造成了网络使用方式僵化、无线频谱规划紧张,难以满足未来网络通信技术智能化、泛在化、融合化的发展趋势。作为认知无线电技术与自组织网络结合的认知自组织网络,采用频谱共享的认知无线电系统,使网络从静态工作模式发展到动态自适应模式,在一定程度上解决了频谱资源短缺的问题,提高了网络对环境的适应能力。然而,在认知自组织网络中,网络拓扑具有时变性,容易形成网络孤岛,进而导致认知节点之间的链路频繁的断开,需要中继节点以多跳存储携带转发、接触关系预测的方式将数据包以动态机会路由模式转发至目的节点。因此,本文采用接触可预测的方法描述不同网络孤岛间SU节点间接触关系,基于接触可预测的认知自组织网络的特性,有针对性地开展了以下研究工作:1、在接触可预测的认知自组织网络可用频谱与网络拓扑时变性的前提下,提出了一种动态多路径机会路由选择算法-SefRoute。通过利用电磁场中电场线的分布,初步确定多条不相交路径的轨迹,结合社交网络中相关知识,构造出节点之间相遇概率的函数,通过机会连接率将多元社交关系中的中心性、相似性与信任度三个重要因素与节点相遇概率结合在一起。结合认知自组织网络自身的特性构造出三个重要因素的相关函数,最终在多条不相交路径轨迹的基础上确定多路径机会路由。仿真结果表明所提SefRoute算法在成功传输率、时延等方面均有很好的性能。2、定义了一种链路稳定度指标的度量方法,结合路径稳定度与业务类型优先级别提出多路径路由动态速率分配算法。考虑用户不同类型业务的性能参数,引入流量感知因子,在保证不同类型业务对流量需求的基础上,将路径稳定性因子考虑在内,利用数学工具-非合作微分博弈建立多路径路由动态的速率分配模型。构造最优问题并得到最优解,仿真分析表明,该算法在不同业务对QoS(服务质量)需求方面有很好的性能。
[Abstract]:With the rapid development of wireless communication technology, the available spectrum resources become increasingly tight, especially with the rapid development of wireless local area network, wireless personal area network and wireless metropolitan area network. The current broadband wireless application technology has been unable to meet the needs of people. The current spectrum allocation system has resulted in rigid network usage and tight wireless spectrum planning, which makes it difficult to meet the development trend of intelligent, ubiquitous and integrated network communication technology in the future. As a combination of cognitive radio technology and ad hoc network, the cognitive ad hoc network uses spectrum sharing cognitive radio system to develop the network from static mode to dynamic adaptive mode. To some extent, the shortage of spectrum resources is solved, and the adaptability of network to environment is improved. However, in cognitive ad hoc networks, the network topology is time-varying, which is easy to form a network island, which leads to frequent disconnection of links between cognitive nodes, which requires relay nodes to be stored and forwarded in multiple hops. Contact relationship prediction forwards packets to the destination node in dynamic opportunity routing mode. Therefore, the contact predictability method is used to describe the contact relationship between SU nodes in different network islands. Based on the characteristics of the contact-predictable cognitive ad hoc networks, the following research works are carried out: 1. In this paper, a dynamic multipath opportunistic routing algorithm, SefRoute., is proposed on the premise of the temporal variability of the available spectrum and the topology of the cognitive ad hoc networks. By using the distribution of electric field lines in the electromagnetic field, the locus of several disjoint paths is preliminarily determined, and the function of the probability of meeting between nodes is constructed by combining the relevant knowledge in the social network, and the centrality of the multivariate social relations is determined by the chance connection rate. Three important factors of similarity and trust are combined with the encounter probability of nodes. Based on the characteristics of cognitive ad hoc networks, the correlation functions of three important factors are constructed. Finally, the opportunity routing of multiple paths is determined on the basis of multiple disjoint paths. The simulation results show that the proposed SefRoute algorithm has good performance in the aspects of successful transmission rate and delay. Combining path stability with traffic type priority, a dynamic rate allocation algorithm for multipath routing is proposed. Considering the performance parameters of different types of services and introducing a traffic awareness factor, the path stability factor is taken into account on the basis of ensuring the traffic requirements of different types of services. A dynamic rate allocation model of multipath routing is established by using non-cooperative differential game, which is a mathematical tool. The optimal problem is constructed and the optimal solution is obtained. The simulation results show that the algorithm has good performance in different traffic requirements for QoS (quality of Service).
【学位授予单位】：河北工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN925

【参考文献】