超密集无线网络内容缓存优化方法研究

发布时间：2018-10-13 13:20

【摘要】：数据业务需求的爆发式增长使超密集无线网络成为研究热点。根据网络中内容受欢迎的程度即流行度将热点内容进行缓存,将通信、计算、存储多维资源进行联合优化,是超密集网络研究的新方向。在超密集部署网络中,将小蜂窝的基站按照地域位置和信号强度等条件进行分簇,如何分簇对超密集网络的切换次数和吞吐量将产生很大的影响,这是超密集研究的另一研究热点。本论文在介绍了超密集无线网络体系架构与关键技术的基础上,主要研究了以下三个方面的内容:(1)提出在超密集网络中将通信、计算、存储多维资源进行联合优化的体系架构,利用分布式存储,使热点内容更接近用户,同时减少网络的通信量;(2)提出了一种超密集无线网络基于内容缓存的能效优化方法,能够帮助一部分节点和用户从中继节点基站分流流量,从而降低能耗。仿真结果表明,与随机缓存方案相比,所提出的基于内容缓存的能效优化方法在能效和内容访问命中率方面都具有良好的性能;(3)在超密集无线网络中,基于马尔科夫模型对移动终端在超密集分簇网络下的移动状态进行建模。仿真结果表明该方法可以有效提高网络系统的吞吐量;同时,在合理设计参数情况下也可以降低系统的阻塞概率。最后,对本论文进行概括总结,并对下一步工作的研究方向和思路进行了展望。
[Abstract]:The explosive growth of data service demand makes the ultra-dense wireless network become a research hotspot. According to the popularity of the content in the network, it is a new direction to cache the hot content and to optimize the communication, computation and storage of multi-dimensional resources. In ultra-dense deployment networks, the base stations of small cells are clustered according to the conditions of location and signal strength. How to cluster will have a great impact on the handover times and throughput of the super-dense networks. This is another hot spot in super-intensive research. Based on the introduction of the architecture and key technologies of the ultra-dense wireless network, this paper mainly studies the following three aspects: (1) the architecture of the communication, computing and storage of multi-dimensional resources in the ultra-dense network is proposed. Distributed storage is used to make hot content closer to users, while reducing network traffic. (2) an energy efficiency optimization method based on content cache is proposed for ultra-dense wireless networks. It can help some nodes and users to divert traffic from the relay node base station, thus reducing energy consumption. Simulation results show that the proposed energy efficiency optimization method based on content cache has good performance in terms of energy efficiency and content access hit rate compared with random cache scheme. (3) in ultra-dense wireless networks, Based on Markov model, the mobile state of mobile terminal in super dense clustering network is modeled. The simulation results show that the proposed method can effectively improve the throughput of the network system and reduce the blocking probability of the system under the condition of reasonable design parameters. Finally, the thesis is summarized, and the research direction and thought of the next work are prospected.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TN929.5

【参考文献】