多网络覆盖下网络选择及业务分配算法研究

发布时间：2018-05-17 21:47

本文选题：网络选择 + 业务分类　；参考：《哈尔滨工业大学》2014年硕士论文

【摘要】：如今，多种不同的无线和有线网络技术的存在，努力满足不同用户的需求和要求。在不久的将来，网络运营商将尝试结合各种技术以建立大型异构网络向用户提供更好的服务。然而，这种网络的设计和开发需要考虑许多问题，如会话控制、授权、认证，服务质量（QoS），收费，用户移动性等。用户对宽带服务的需求增加，至少在热点地区，未来的无线通信系统将考虑提供多种无线网络的集成系统。这种异构网络的集成将具有特定的高数据速率和未经授权的优势频谱。因此，异构网络的网络选择技术在确保服务质量上起着至关重要的作用。对于用户来说，通过异构网络融合，可以根据业务需求和网络状态等因素随时选择接入到最合适的网络中，已达到满足用户自身灵活多变的个性化业务需求的目的。在异构全IP无线网络环境中的服务承载要求选择一个最优的访问网络。选择最优的网络将导致成本上升等不良影响或糟糕的服务体验。网络选择在这样一个环境中受到多个因素的影响，目前一个没有完整的解决方案来解决这个问题。本课题在多网络覆盖环境中，针对每种接入网络提供的性能（主要包括：链路质量、可靠性、网络传输时延、带宽、服务费用、移动性支持等），研究多模终端如何根据各接入网络特性并结合网络负载状态和业务需求，选择用户最满意的网络进行业务承载。本文设计了一个基于多网络的网络选择方案，该方案的设计目标是在任何时候都为用户提供可用的最好的QoS服务质量。本文首先对异构网络以及3GPP定义的四种业务类型进行研究，基于此研究基础提出的网络选择方案包括两个部分，第一部分利用多属性的层次分析法（AHP）来决定用户服务类型QoS的相对权重，针对不同类型的业务进行QoS分析。第二部分采用理想值近似性排序算法（TOPSIS）进行候选网络排序来确定最优承载网络。本文提出的基于业务分类的网络选择方案克服了传统平均权值方法不考虑业务之间的QoS需求差别，集中选择QoS最优网络进行业务承载，从而无法合理、有效地平衡各种候选网络的负载，，造成网络拥塞以及资源浪费的问题。仿真分析表明本文提出的方案可实现将流媒体类和交换类业务更多地切换到WiMAX这种相对高数据率、低误码率的网络中，从而克服了传统平均权值方法将会话类和流类业务大量接入UMTS网络，而将交换类和后台类业务更多接入WLAN网络而造成的UMTS网络长期阻塞，降低终端接入UMTS的成功率的问题。本课题设计的方案可以合理选择目标网络，同时有效平衡多个候选网络的负载，从而达到有效利用网络资源的目的。
[Abstract]:Nowadays, many different wireless and wired network technologies exist to meet the needs and requirements of different users. In the near future, network operators will try to combine various technologies to build large heterogeneous networks to provide better services to users. However, many problems need to be considered in the design and development of this network, such as session control, authorization, authentication, QoS QoS, fees, user mobility, and so on. Users' demand for broadband services is increasing. At least in hot spots, future wireless communication systems will consider integrated systems that provide multiple wireless networks. The integration of such heterogeneous networks will have specific high data rates and unauthorized dominant spectrum. Therefore, the network selection technology of heterogeneous networks plays an important role in ensuring QoS. For users, through heterogeneous network fusion, they can be accessed to the most suitable network at any time according to the factors such as service requirements and network status, which has met the user's own flexible and changeable needs of personalized business. In heterogeneous all IP wireless network environment, the service load needs to choose an optimal access network. Choosing the best network will lead to adverse effects, such as rising costs, or poor service experiences. The network choice is affected by many factors in such an environment, and there is no complete solution to solve this problem. This paper aims at the performance of each access network in multi-network coverage environment (including link quality, reliability, network transmission delay, bandwidth, service cost, etc. Based on the characteristics of each access network and combined with the network load state and the service requirement, the paper studies how to select the most satisfied network to carry the service. This paper designs a multi-network-based network selection scheme, which aims to provide the best QoS quality of service available to users at all times. In this paper, we first study the heterogeneous network and the four types of services defined by 3GPP. The network selection scheme based on this research includes two parts. In the first part, the relative weight of user service type QoS is determined by using multi-attribute analytic hierarchy process (AHP), and the QoS analysis for different types of services is carried out. In the second part, the ideal value approximation sorting algorithm (Topos) is used to sort the candidate networks to determine the optimal bearer networks. The network selection scheme based on service classification in this paper overcomes the traditional average weight method without considering the difference of QoS requirements between services, and centralizes the selection of QoS optimal network for service carrying, so it is not reasonable. Effectively balancing the load of various candidate networks, resulting in network congestion and waste of resources. Simulation results show that the proposed scheme can switch more streaming and switching services to WiMAX networks with relatively high data rate and low bit error rate. This method overcomes the problem that the traditional average weight method connects session and stream services to UMTS network in large quantities, and reduces the success rate of terminal access to UMTS because more switching and back-end services are connected to WLAN network for a long time. The scheme designed in this paper can select the target network reasonably and balance the load of several candidate networks effectively, so as to make effective use of network resources.
【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】：2014
【分类号】：TN92

【相似文献】