基于软件定义无线网络的云业务上行调度方案

发布时间：2018-04-17 12:17

本文选题：软件定义无线网络 + 无线云计算　；参考：《南京邮电大学》2016年硕士论文

【摘要】：随着移动通信技术的快速发展,各式各样的新兴业务不断涌现,这对于通信网络的性能提出了越来越高的要求,比如被视为未来信息技术变革核心的无线云计算业务,其对网络性能的需求不仅仅在于一定的传送速率,更需要使终端的发送能耗最小,以延长电池寿命。然而在现有的LTE网络结构下,由于控制和转发的紧耦合,不同的网络节点有各自独立的控制系统,在此基础上实行的分类调度方案只能将云业务归为某类业务进行笼统地处理,并不能针对其特点进行灵活地配置,因而不能很好地满足以无线云计算为代表的新兴业务对网络性能的个性化需求。因此本文在软件定义无线网络这一新的网络架构下,研究了基于应用的Qo S控制方法,并在此方法的基础上提出了针对单云用户的云业务上行调度方案,该方案首先将资源分配问题归纳成二进制整数规划模型进行求解,然后利用动态规划的方法得出在当前的信道条件下最佳的发送速率,最后通过基于应用的Qo S控制方法调节云用户的发送速率来使其发送能耗最小,以在满足多用户速率要求的同时,保障云业务对网络性能的特殊需求。此外,本文在单云业务的上行调度方案的基础上,提出了SDWN下的多云业务上行调度方案,该方案不仅能在保障多个普通业务速率需求的同时,对并发的多个云业务进行速率调整,使其能够随着各自信道增益的变化而变化来优化终端能耗,而且引入了代表用户满意度的效用函数,使资源能更加合理、公平地进行分配,最大化所有用户的满意度之和并保障业务间的相对公平性。仿真结果表明,针对单云用户和多云用户的云业务上行调度方案,都能有效减小云用户的能量消耗,在保障网络整体性能的同时达到满足其特殊QoS的目的。
[Abstract]:With the rapid development of mobile communication technology, a variety of emerging services have been emerging, which put forward more and more high requirements for the performance of communication networks, such as wireless cloud computing business, which is regarded as the core of information technology reform in the future.The demand for network performance lies not only in a certain transmission rate, but also in minimizing the transmission energy consumption of the terminal, so as to prolong the battery life.However, under the existing LTE network structure, due to the tight coupling between control and forwarding, different network nodes have their own independent control systems. On this basis, the classified scheduling scheme can only classify cloud services into certain types of services for general processing.It can not be configured flexibly according to its characteristics, so it can not meet the personalized demand of network performance of the emerging business represented by wireless cloud computing.Therefore, under the new network architecture of software definition wireless network, this paper studies the QoS control method based on application, and puts forward a cloud service uplink scheduling scheme for single cloud users on the basis of this method.In this scheme, the problem of resource allocation is first solved by binary integer programming model, and then the optimal transmission rate under the current channel condition is obtained by using dynamic programming method.Finally, the transmission energy consumption is minimized by adjusting the transmission rate of cloud users through the QoS control method based on application, so that the special requirements of cloud services for network performance can be guaranteed while satisfying the requirements of multi-user rates.In addition, on the basis of the uplink scheduling scheme of single cloud service, this paper proposes an uplink scheduling scheme for cloudy services under SDWN.The rate of concurrent cloud services is adjusted to optimize the terminal energy consumption with the change of channel gain, and the utility function representing user satisfaction is introduced so that the resources can be allocated more reasonably and equitably.Maximize the sum of customer satisfaction and ensure relative fairness between services.The simulation results show that the upstream scheduling scheme of cloud services for both single and cloudy users can effectively reduce the energy consumption of cloud users and achieve the purpose of satisfying their special QoS while ensuring the overall performance of the network.
【学位授予单位】：南京邮电大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TN929.5

【相似文献】