【作者】 马学彬；

【导师】 温涛；

【作者基本信息】 东北大学 ， 计算机应用技术， 2009， 博士

【摘要】 IETF提出了移动IPv6协议用来支持移动用户无缝漫游Internet。移动IPv6是目前移动通信方向的研究热点,并将在未来无线通信体系中占有重要地位。但是,现有的移动IPv6在切换技术方面还不成熟,存在切换信令开销大和切换延迟高的不足。另外,随着互联网中视频、语音、图像等多媒体应用的快速发展,为移动IPv6网络提供服务质量和组播通信方面的支持也成为人们所关心的问题。因此对移动IPv6中切换技术、服务质量保证机制和组播技术的研究不仅具有重要的理论意义,而且具有广阔的应用前景。本文阐述了移动IPv6的发展背景和研究现状,分析了现有移动IPv6技术的优缺点,在此基础上,围绕移动IPv6中切换、服务质量和组播技术三个方面进行了研究,并给出了相应的解决方法,主要研究成果如下(1)针对层次移动IPv6中锚点域固定不变从而使移动锚点容易成为失效集中点和网络瓶颈的问题,提出一种基于动态域的移动IPv6切换方案,在该方案中移动锚点可以由移动节点动态选择,所以对于一个移动节点而言,任何一个接入路由器都有可能成为它的移动锚点,并且每个锚点域是动态变化的。在一个锚点域内,如果移动节点只是改变当前的位置,则无需向家乡代理和通信对端注册,只要向区域锚点注册其在线转交地址即可。另外在该方案中提出一种基于锚点饱和度的动态调整接纳阈值的方法,有效减少了在移动锚点处发生拥塞的几率,使系统中各节点的负载达到均衡。(2)在移动IPv6服务质量保证机制方面,提出一种移动IPv6中的QoS上下文转移方案,该方案可以在移动节点执行切换的同时为实时应用提供服务质量保证。由于采用了基于快速层次移动IPv6框架,减少了移动节点切换期间的信令开销和延时。同时,该方案可以减少移动节点切换后重新发起信令建立QoS时由转发处理所带来的延时,从而减少了移动节点的QoS中断时间。性能分析和仿真结果表明该方案可以显著降低实时业务切换时的延迟抖动,特别是当移动节点进行乒乓切换时有较好的表现,实现了移动节点的平滑切换。另外,针对移动节点有可能不支持任何一种移动性管理协议的问题,提出一种具有QOS保证机制的PMIPv6协议。该协议切换简单、快速,不需要移动节点在IP层参与,并且能有效保证实时业务的服务质量要求。实验结果表明,该协议可以有效降低实时业务切换时的延迟和开销,并且有较低的丢包率,从而满足了移动节点对实时应用的要求。(3)分析了已有移动组播协议在实际应用中的不足,提出一种基于移动预测的快速分层移动组播体系结构。该方案通过移动预测,在切换前为移动节点配置好切换信息,并且在切换时使用基于FMIPv6改进的快速组播切换方法,有效减少了切换延迟以及由此引发的组播数据分组丢失。另外,由于采用层次型结构的移动组播管理,屏蔽了移动节点在子网内的移动,减少了因移动节点移动而重构组播树的频率最后,对全文进行了总结,并对有待于进一步研究的问题进行了展望。更多还原

【Abstract】 Mobile IPv6 (MIPv6) is a protocol proposed by IETF to provide support to mobile users for seamless roaming on the Internet. Mobile IPv6 is presently a hot research subject in mobile communication, and will play an important role in the wireless communication system of future. However, MIPv6 is still immature in the handover technology such as high handover signaling overhead and long handover delay. Besides, with the rapid development of multimedia on the internet-video, voice, image, etc., support of Quality of Service (QoS) and multicast for MIPv6 network is also concerned by people. Therefore, research on handover technology, QoS assurance mechanism and multicast technology of MIPv6 not only has great theoretical significance, but also enjoys good prospect in application.This dissertation expounds the research background and current situation of MIPv6 and analyzes the advantages and disadvantages of its handover technology. Based on this, this dissertation concentrates on the handover technology, QoS and multicast in MIPv6 and proposes corresponding solutions. The main achievements are as follows:(1) A dynamic domain based handover scheme for mobile IPv6 is proposed, for Mobile anchor point (MAP) may become a single point or network bottlenecks because anchor point domain is unchanged in HMIPv6. In this scheme, mobile nodes (MN) could dynamically select mobile anchor point, so any access point (AP) could become its MAP and each MAP domain is dynamic. If a MN only changes its current position in a MAP domain, it only needs to register its on-link care-of address with the MAP rather than the home agent (HA) and corresponding node (CN). In addition, this scheme proposes a method of changing acceptation threshold dynamically which is based on the saturation of MAP. It reduces the congestion probability of MAP and makes the load balance on every node.(2) A QoS context transfer scheme for MIPv6 is proposed in the QoS assurance mechanism. This scheme could guarantee the quality of service (QoS) for real-time applications when Mobile Node performs handover. Because this method adopts the fast handover for hierarchical MIPv6 architecture, the signaling overhead and handover delay is reduced. Meanwhile, the scheme could reduce the delay caused by forwarding process after Mobile Node is switched and QoS is reestablished by reinitiating QoS signal, thereby reducing the QoS disruption. Performance analysis and simulations show that this scheme can reduce the delay and jitter obviously and can perform better especially in ping-pong movement, consequently realizing smooth handover. Besides, a PMIPv6 protocol with QoS assurance mechanism is proposed for the problem that MN may not support any kind of mobility management protocol. This protocol makes the handover simpler, faster and doesn’t need the MN participate in IP layer, and can effectively guarantee the QoS requirements of real-time business. Simulations show that the proposed protocol could meet the needs of real-time applications, because it reduces the delay and overhead of handover obviously, and has lower packet loss ratio.(3) A mobility prediction-based fast and hierarchical mobile multicast architecture is proposed on the basis of the analysis of the limitations in existing mobile multicast protocols when they are applied into real mobile networks. This scheme could configure the handover information before multicast handover by mobility prediction, and could reduce handover delay and subsequent packet loss rate with improved FMIPv6 handover method when mobile nodes are playing handover. In addition, it isolates the subnet movements from outside, and reduces the frequency of rebuilding multicast tree because of mobile nodes’movements by using a hierarchical architecture.Finally, a summary of the dissertation is made and some problems remained to be further researched are discussed.更多还原