移动互联网环境下基于模糊神经网络的异构网络选择机制研究

发布时间：2018-05-12 13:40

本文选题：移动互联网 + 模糊神经网络　；参考：《北京邮电大学》2014年硕士论文

【摘要】：随着移动通信、移动互联网等技术的越来越高速的发展,网络已经成为社会的基础设施,它有着泛在性和随机接入性的特征,推动着异构无线接入技术的网络相互融合、协作共存,逐渐成为未来无线网络的必然发展趋势。在多种无线接入技术共存的异构网络环境中,垂直切换渐渐取代了传统的水平切换,成为异构网络选择机制的基础,因此对垂直切换机制的研究对均衡网络负载,为用户提供高质量且可靠的服务具有极其深远的意义。但是目前的垂直切换算法仅仅单一的集中于多属性决策、成本和策略等,没有很好的统筹整个网络的资源,亦不能对所有判决参数进行精确的建模。基于此,为了适应多种接入技术的异构网络环境的变化,需要增加人工修正调整的过程,我们在无线网络资源协同管理的框架下,提出基于模糊逻辑理论和人工神经网络的改进算法,能够使上述存在的问题得到一定程度的改善,可以作为实现异构网络切换选择的较好机制。本文的研究工作主要包括如下几方面： (1)提出异构无线接入网络环境下的无线网络资源协同管理的创新机制,为保证网络的负载和服务质量,采用模糊逻辑理论和人工神经网络相结合的算法。 (2)模拟不同无线接入技术的多小区的场景部署,并进行讨论。该场景包括三种网络：通用移动通信系统(UMTS)、GSM/EDGE无线接入网(GERAN)、无线局域网(WLAN)。 (3)提出的模糊神经无线网络资源协同管理算法,能够协同管理可用的无线网络资源,具体分为两个步骤。第一步是在三种无线接入技术存在的网络中,选择合适的小区组合,第二步是在该小区组合中选择最适合用户接入的无线技术,并分配给用户适当的带宽。 (4)对不同的实现进行仿真和对比,仿真结果表明提出的模糊神经方法能够处理异构网络环境的复杂性和不确定性,由此可见该算法是一种很有前途的选择机制。 (5)此外,该算法还通过多目标决策机制,考虑到运营商的不同策略以及用户的不同的主观标准,如均衡不同无线接入技术间的业务流量,或赋予某种无线接入技术以优先级。
[Abstract]:With the rapid development of mobile communication, mobile Internet and other technologies, network has become the infrastructure of society. It has the characteristics of universality and random access, and promotes the integration of heterogeneous wireless access technology. Cooperative coexistence has gradually become the inevitable development trend of wireless networks in the future. In the heterogeneous network environment where multiple wireless access technologies coexist, vertical handoff gradually replaces the traditional horizontal handoff and becomes the basis of heterogeneous network selection mechanism, so the research on vertical handoff mechanism is to balance the network load. Providing high quality and reliable services to users is of great significance. However, the current vertical handoff algorithm is only focused on multi-attribute decision making, cost and strategy, and so on. It does not well coordinate the resources of the whole network, nor can it accurately model all the decision parameters. Based on this, in order to adapt to the change of heterogeneous network environment of multiple access technologies, we need to increase the process of manual correction and adjustment. We use the framework of cooperative management of wireless network resources. An improved algorithm based on fuzzy logic theory and artificial neural network is proposed, which can improve the problems mentioned above to a certain extent and can be used as a better mechanism to realize handoff selection in heterogeneous networks. The research work of this paper mainly includes the following aspects: In order to ensure the load and quality of service of the network, fuzzy logic theory and artificial neural network are used to combine the fuzzy logic theory with the artificial neural network. Simulation and discussion of multi-cell scenario deployment for different wireless access technologies. The scenario consists of three types of networks: the Universal Mobile Communication system (UMTS), the GSM / edge wireless access network, the WLAN, and the WLAN. The fuzzy neural network resource collaborative management algorithm, which can manage the available wireless network resources in cooperation, is divided into two steps. The first step is to select the appropriate cell combination in the network where the three wireless access technologies exist. The second step is to select the most suitable wireless technology for the user access in the cell combination and allocate the appropriate bandwidth to the user. Simulation results show that the proposed fuzzy neural method can deal with the complexity and uncertainty of heterogeneous network environment, so this algorithm is a promising choice mechanism. In addition, the algorithm takes into account different strategies of operators and different subjective standards of users, such as balancing traffic between different wireless access technologies, or giving priority to certain wireless access technologies.
【学位授予单位】：北京邮电大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP18;TN929.5;TP393.01

【参考文献】