EARS: CogCloud: A Spectrum-Efficient and Green Cloud Platform for Radio-As-A-Service Over a Cognitive Radio Substrate

虚拟化在云计算中得到了广泛的应用，可以实现多个租户之间的性能隔离，提高资源利用率和能源效率。然而，对无线虚拟化的研究仍处于起步阶段。由于无线网络在过去二十年中的快速增长，以及大部分主要无线频谱已经分配给特定的服务提供商或专门使用的服务，频谱的稀缺性已经成为一个严重的问题。另一方面，最近的测量表明，许可频谱严重未得到充分利用。这导致了对认知无线电传输策略的极大兴趣。该项目的目标是开发一个频谱高效的云平台，即CogCloud，以在认知无线电基板上实现无线电即服务（RaaS），并通过设计一个两级闭环控制框架来最小化其能耗，该框架利用云级和bs级优化分别对无线电资源和无线用户进行粗粒度和细粒度控制。CogCloud预计具有以下理想特性：1)多个独立移动虚拟网络运营商（mvno）的RaaS：认知无线电基板中的无线电资源作为服务提供给多个独立的mvno；2)性能隔离：一个MVNO的变化（例如无线用户数量、流量负载等）不会影响其他MVNO的无线用户的性能；3)频谱效率：频谱可用性被认知无线电充分利用，并由云有效管理，以支持qos无线通信；4)能源效率：认知无线电基质及其无线电资源以最节能的方式运行。为了实现这些目标，该项目被组织成四个有凝聚力的研究重点：1)系统架构设计和实现；2)云级优化；3) bs级优化，实现细粒度控制；4)验证与性能评价。该项目预计将对云计算、认知无线网络和绿色通信的发展产生重大影响，并通过出版物、研讨会、讲习班和国际合作，促进公众对新兴研究领域的理解，如无线虚拟化和基于云的无线接入网络（ran）。为吸引来自代表性不足群体的学生作出了特别努力。

Virtualization has been widely used in cloud computing to achieve performance isolation among multiple tenants, and to improve resource utilization and energy efficiency. However, research on wireless virtualization is still in its infancy. Due to the rapid growth in wireless networks in the last two decades and the fact that much of the prime wireless spectrum has already been allocated for specific service providers or services for exclusive use, the scarcity in the spectrum has become a serious concern. On the other hand, recent measurements have shown that the licensed spectrum is severely under-utilized. This has led to significant interest in cognitive radio transmission strategies.The objective of this project is to develop a spectrum-efficient cloud platform, namely CogCloud, to enable Radio-as-a-Service (RaaS) over a cognitive radio substrate, and minimize its energy consumption by designing a two-level closed-loop control framework that leverages cloud-level and BS-level optimization for coarse-grained and fine-grained control respectively over radio resources and wireless users. CogCloud is expected to have the following desirable properties: 1) RaaS for multiple independent Mobile Virtual Network Operators (MVNOs): Radio resources in a cognitive radio substrate are provided as a service to multiple independent MVNOs; 2) performance isolation: Changes in an MVNO (such as the number of wireless users, their traffic load, etc) do not affect the performance of wireless users of other MVNOs; 3) spectrum efficiency: spectrum availability is fully exploited by cognitive radios and efficiently managed by the cloud to support QoS-enabled wireless communications; 4) energy efficiency: the cognitive radio substrate and its radio resources are operated in the most energy-efficient way. To achieve these goals, the project is organized into four cohesive research thrusts: 1) system architecture design and implementation; 2) cloud-level optimization; 3) BS-level optimization for fine-grained control; and 4) validation and performance evaluation.The project is expected to make a significant impact on the advancement of cloud computing, cognitive radio networking and green communications, and advance public understanding of emerging research areas, such as wireless virtualization and cloud-based radio access networks (RANs), via publications, seminars, workshops and international collaborations. Special efforts are made to engage students from under-represented groups.