CAREER: Programmable Mobile Networking

移动应用基础、移动服务需求和底层网络技术的快速发展给移动计算和通信研究界带来了巨大的挑战。现有的移动网络体系结构表现出三个基本限制，阻碍我们迎接这一挑战：(I)缺乏对音频、视频和实时数据传输的服务质量支持；(Ii)在适应新的移动用户和服务提供商需求方面缺乏内在的体系结构灵活性；以及(Iii)缺乏实现和部署新的和独特的移动网络体系结构的自动化过程。针对这些限制，PI认为，迫切需要提出、调查和评估现有移动网络架构的替代方案。拟议的研究远远超出了对另一种实验性移动网络体系结构的建议，例如移动IP++。相反，它呼吁采用新的创新方法来设计、开发、部署、观察和分析下一代移动网络体系结构，以响应移动用户、服务提供商和设备供应商的未来需求和要求。这项研究的目标是：(I)为虚拟移动网络及其体系结构的生成和原型建立方法和创建环境；(Ii)为一组不同的基线体系结构部署可编程基础服务、网络控制器和开放可编程网络接口，其中将包括新的宽带蜂窝网络体系结构、下一代移动互联网体系结构和移动自组织网络体系结构；(Iii)在同一物理高速移动网络试验床上演示这些基准虚拟移动网络的产生、可视化和管理；以及(Iv)通过可视化、管理和改进来研究所产生的体系结构的架构设计空间的权衡。实现能够剖析、产生、可视化和管理不同的虚拟移动网络体系结构的衍生网络将代表网络架构师设计、部署和管理新的和新颖的移动网络体系结构的方式的重大转变。PI还通过在教育中引入新的和创新的举措来回应工业界和学术界对具有移动网络和分布式系统技能的研究生的强烈需求。PI已经开设了一门无线和移动网络的研究生课程，并正在教授另一门新开设的分布式电信系统课程。每门课程通过介绍一些行业领先研究人员的最先进的研究研讨会来阐述基本概念，并强调“动手”学期项目。该协会非常热衷于促进学术界和产业界之间的互动。这种互动提供了一个极好的学习机会，使学生、教师和行业都受益。

The rapidly evolving nature of the mobile application base, mobile service demands and underlying network technologies presents a significant challenge to the mobile computing and communications research community. Existing mobile network architectures exhibit three basic limitations that prevent us from meeting this challenge: (i) the lack of quality of service support for the transport of audio, video andreal-time data; (ii) the lack of intrinsic architectural flexibility in the adaptation to new mobile user and service provider needs; and (iii) the lack of automation of the process of realization and deployment ofnew and distinct mobile network architectures. In response to these limitations, the PI argues that there is a critical need to propose, investigate and evaluate alternative mobile network architectures tothose existing today. The proposed research goes far beyond a proposal for yet another experimental mobile network architecture, e.g., mobile IP++. Rather, it calls for new innovative approaches to the way wedesign, develop, deploy, observe and analyze next-generation mobile networking architectures in response to the future needs and requirements of mobile users, service providers and equipment vendors.The goal of this research in "spawning networks" is to: (i) establish a methodology and creation environment for spawning and prototyping virtual mobile networks and their architectures; (ii) deploy programmable foundation services, network controllers and open programmable networking interfaces for a distinct set of baseline architectures that will include a new broadband cellular network architecture, next-generation mobile Internet architecture and mobile ad hoc network architecture; (iii) demonstrate the spawning, visualization and management of these baseline virtual mobile networks over the same physical high-speed mobile networking testbed; and (iv) investigate trade-offs in the architectural design space of the spawned architecture through visualization, management and refinement.The realization of spawning networks capable of profiling, spawning, visualizing and managing distinct virtual mobile networking architectures will represent a major shift in the way network architects design, deploy and manage new and novel mobile networking architectures.The PI is also responding to the strong demands from industry and academia for graduate students with mobile networking and distributed systems skills through the introduction of new and innovative initiatives in education. The PI has introduced a graduate-level course in Wireless and Mobile Networking and is teaching another newly created course in Distributed Telecommunications Systems. Each course addressesfundamental concepts, through the presentation of state-of-the-art research seminars from some of industry's leading researchers and emphasizes "hands on" term projects. The PI is very keen to promoteinteraction between academia and industry. This interaction provides a great learning opportunity benefiting students, faculty and industry alike.