Software Defined Cellular Wireless Networking

软件定义蜂窝无线网络 软件定义网络（SDN）是一项新的技术进步，旨在将控制平面与转发数据包的数据平面物理分离，以提高网络的可编程性，可扩展性和性能，以及系统的易管理性。为了更好地支持在移动的蜂窝系统中生成和传输的数据量的快速增长，采用SDN已经成为必要，SDN带来了网络管理和控制的逻辑集中的概念。用户的需求也从基本的语音和短消息服务转变为高质量的音频和视频流。物联网的发展预计也会增加蜂窝系统的压力。采用所提出的SDN架构不仅可以增加吞吐量，还可以减少系统中的延迟。 本研究项目考察了SDN引入的架构应用于蜂窝系统的进步效果，并探讨了这种发展的可靠性和安全性。当前蜂窝模型中的主要组件（诸如分组数据网络网关（P-GW））过于专业化和昂贵，并且几乎没有为添加新服务提供空间。拟议的蜂窝SDN提供了很大的灵活性，因为可以部署应用程序编程接口，以更快、更低的成本添加新服务。通用SDN架构在蜂窝系统中不可实现，因为在P-GW中的数据平面和控制平面之间没有明确的分离。SDN架构由三个基本层组成：基础设施层，包括网络元素的数据平面，控制层和应用层。OpenFlow是用于控制平面和数据平面之间通信的最流行的协议。 我们的研究将仔细检查通用SDN模型，旨在开发框架，使其更具体和可实施的移动的蜂窝系统。我们正在考虑SDN控制器的不同部署模型及其对端节点服务质量的影响。系统的可靠性和安全性是系统的关键方面，但它没有得到太多的关注。本研究将分析SDN引入的逻辑集中和可编程性的安全含义。 几位专家将在这个研究项目中接受培训，这些专家将成为加拿大行业的蜂窝SDN领导者。这项研究的结果将是一个重要的工具，在提高可扩展性，性能和易于管理的系统，以及消除供应商锁定与当今的蜂窝系统。系统开发人员和行业运营商将利用研究结果来提高其系统的响应能力。加拿大运营商将受益于蜂窝SDN提供的可扩展性、改进的性能和易于管理的特性。

Software Defined Cellular Wireless Networking Software Defined Networking (SDN) is a new technology advancement aimed at physically separating the control plane from the data plane which forwards packets, in order to improve programmability, scalability and performance of the network, as well as ease of management of the system. It has become necessary to adopt SDN which brings about the concept of logical centralization of network management and control in order to better support rapid growth in volume of data generated and transmitted in mobile cellular system. The needs of users has also changed from basic voice and short message services to high quality audio and video streaming. Evolution of Internet of Things is expected to increase the strain on the cellular systems as well. Adoption of proposed SDN architecture will not only increase throughput but also reduce latency in the system. This research program examines the effect of advancement in architecture introduced by SDN as it applies to cellular systems, and explores the reliability and security of this development. Major components in the current cellular model such as the packet data network gateway (P-GW) are too specialized and expensive, and offer little room for addition of new services. Proposed cellular SDN offers a lot of flexibility as application programming interfaces can be deployed to add new services quicker and at much lower cost. The generic SDN architecture is not implementable in cellular systems as there is no clear separation between data and control planes in P-GW. SDN architecture consists of three basic layers: the infrastructure layer which encompasses the data planes of network elements, the control layer and the application layer. OpenFlow is the most popular protocol used for communication between the control and data planes. Our research will scrutinize the generic SDN model with aim of developing the framework to make it more specific and implementable in mobile cellular systems. We are considering different deployment models for the SDN controller and their effects on the quality of service experienced at the end nodes. Reliability and security of the system are key aspects of the system but it has not received much attention. This research will analyze the security implications of logical centralization and programmability introduced by SDN. Several experts will be trained in this research program which will become cellular SDN leaders in Canadian industry. The results of this research will be an important tool in improving scalability, performance and ease of management of the system, as well as eliminating vendor lock-in associated with today's cellular systems. System developers and industry operators will leverage on the outcome of the research to improve responsiveness of their systems. Canadian operators will benefit from scalability, improved performance and ease of management offered by cellular SDN.