Bringing the Cloud to the Edge in FiWi Enhanced Networks

为了应对移动数据流量前所未有的增长，我们最近展示了通过基于低成本以数据为中心的以太网技术统一以覆盖为中心的4G移动网络和以容量为中心的光纤无线(FiWi)宽带接入网络而获得的显著延迟和可靠性性能提升，并特别关注回程延迟和可靠性对性能的限制影响。FiWi增强型4G移动网络可以被视为朝着满足极低延迟和超高可靠性的要求迈出的重要一步，同时考虑到经济因素并遵循未来5G移动网络的综合愿景。在我们之前的工作中，鉴于去中心化是5G愿景的另一个重要方面，我们还关注了通过在设备端而不是基础设施端利用智能来为FiWi增强型移动网络开发去中心化路由算法。随着互联网向物联网转型，所谓的边缘服务预计将变得越来越重要。行业参与者越来越意识到，通过将当前未经修改的云扩展到基于新兴移动边缘计算功能的分散式两级云-云小程序架构，从而实现计算密集型和延迟敏感型新应用，例如认知辅助、增强现实或触觉互联网应用，例如用于新兴云机器人的人脸识别和导航，从而获得移动-云融合的好处。最近，可以观察到类似的分散化趋势，因为移动基站角色的变化和它成为服务创造中心的诱人机会，以便通过提供丰富的移动宽带体验来盈利，该体验的特点是低延迟、情景和高度个性化的不可预见的应用和服务，有助于刺激创新、创造收入和提高我们日常生活的质量。本研究方案探讨了将Cloudlet和FiWi增强型网络与高级移动边缘计算功能相集成所带来的挑战和机遇。它调查了它们在引领未来触觉互联网方面的工具作用，特别关注突破性的支持云的联网机器人服务和应用。从长远来看，拟议的研究活动旨在利用FiWi支持的人与机器人之间的通信来帮助合并和重组移动互联网、知识工作自动化、物联网、云技术和先进机器人技术，这些技术代表了预计在2025年具有最高潜在经济影响的五项技术，以扩大其对社会的有益影响。**

To cope with the unprecedented growth of mobile data traffic, we recently have demonstrated the significant latency and reliability performance gains obtained from unifying coverage-centric 4G mobile networks and capacity-centric fiber-wireless (FiWi) broadband access networks based on low-cost data-centric Ethernet technologies, paying particular attention to the performance-limiting impact of backhaul latency and reliability. FiWi enhanced 4G mobile networks may be viewed as an important step towards meeting the requirements of very low latency and ultra-high reliability while taking economic considerations into account and following the integrative vision of future 5G mobile networks. In our previous work, we also paid attention to the development of a decentralized routing algorithm for FiWi enhanced mobile networks by exploiting intelligence at the device rather than infrastructure side given that decentralization is another important aspect of the 5G vision. So-called edge services are expected to become increasingly important as the Internet transforms into the Internet of Things. There is a growing awareness among industry players of reaping the benefits of mobile-cloud convergence by extending today's unmodified cloud to a decentralized two-level cloud-cloudlet architecture based on emerging mobile-edge computing capabilities, thus enabling new applications that are both compute intensive and latency sensitive, e.g., cognitive assistance, augmented reality, or tactile Internet applications such as face recognition and navigation for emerging cloud robotics. Recently, a similar trend toward decentralization can be observed for the changing role of the mobile base station and the compelling opportunity for it to become a hub of service creation in order to monetize from the provision of an enriched mobile broadband experience that is characterized by low-latency, contextualized, and highly personalized unforeseen applications and services that help stimulate innovation, generate revenue, and improve the quality of our every-day lives. This research proposal explores the challenges and opportunities arising from the integration of cloudlets and FiWi enhanced networks with advanced mobile-edge computing capabilities. It investigates their instrumental role in ushering in the future Tactile Internet with a particular focus on groundbreaking cloud-enabled networked robotic services and applications. In a long-term vision, the proposed research activities aim at exploiting FiWi enabled human-to-robot communications to help merge and recombine the mobile Internet, automation of knowledge work, Internet of Things, cloud technology, and advanced robotics, which represent the five technologies with the highest estimated potential economic impact in 2025, in order to multiply their beneficial impact on society. **