Towards future mobile networks from an operational perspective

我的研究工作重点是未来的移动网络，以及此类网络成功运营所涉及的挑战。在这方面，我的工作分为三大支柱。第一大支柱侧重于移动网络核心的新颖设计。在移动网络中，核心负责管理和促进对网络的访问。研究表明，由于互联设备数量的增加和设备模式的变化，尤其是物联网(IoT)设备的连接，核心的压力越来越大。在之前的工作中，我已经演示了在云中托管核心具有解决这些问题的潜力。这一点通过一种新颖的、基于云的核心设计得到了证明，该核心仍然与标准无线接入网络(RAN)接口存在问题。然而，由于这项工作的新颖性，网络运营商能否成功采用这种截然不同的模式仍然是一个问题。仍未回答的关键问题包括：转移到云端是否会影响核心的安全性，Open-RAN倡议开创的多供应商方法是否可以适应基于云的核心，以及核心的弹性和可升级性是否可以在云环境中得到改善。我将从移动网络运营商的角度，通过研究核心设计在实际运营环境中的性能、安全性、弹性和可升级性来回答这些问题。这项研究将导致对新型核心设计的修改，以及对部署在云环境中的其他关键基础设施有益的前景。我工作的第二个支柱将重点从核心扩展到RAN。现有移动设备在不同网络之间无缝切换的能力有限。这包括使用相同无线电接入技术(RAT)的网络，例如在公共和专用5G网络之间移动，以及多RAT网络(例如，从5G移动到WiFi)。RAT(WiFi、4G、5G、基于光的Li-Fi以及Starlink和OneWeb等新的非地面技术)的日益普及导致需要一种新的方法来管理不同网络之间的无缝漫游。关键问题包括如何从基于标准的方法采用无缝漫游，网络的哪些部分需要修改以实现这一目标(在移动网络运营商端，或在用户端，或两者都有)，以及需要哪些额外的技术来确保这一愿景。我将从移动网络运营商的角度回答这个问题，与我在移动网络核心方面的工作一样，重点放在运营方面，包括潜在解决方案的部署和管理。这项研究将通过(A)研究现有漫游技术和(B)设计和评估新的漫游技术，以解决现有解决方案的局限性。规划中的园区规模专用5G网络，与现有的园区规模WiFi部署相结合，将帮助研究和评估多RAT网络，而单RAT网络将通过公共5G网络和专用5G部署之间的漫游进行研究。我工作的最后一个支柱专注于开发研究工具和基础设施，将用于实现前两个支柱。软件工具包括开发、测试和评估核心网络所需的框架，以及模拟漫游场景所需的模拟真实世界网络的工具。物理基础设施包括实验室规模和园区规模的移动网络。关键问题包括：研究移动核心网络和漫游需要哪些工具，应如何设置物理基础设施以最好地支持这项研究，以及应如何传播从基础设施中获得的工具和知识以最好地支持未来的研究、其他研究和重复性

My research work focuses on future mobile networks, and the challenges involved in the successful operation of such networks. In this regard, my work is structured into three "pillars".The first pillar focuses on the novel design of the core of mobile networks. In a mobile network, the core is responsible for managing and facilitating access to the network. Studies have demonstrated the increasing strain on the core due to the increase in the number of connected devices and changing device patterns, particularly those of Internet of Things (IoT) devices. In previous work I have demonstrated that hosting the core in the cloud has the potential to solve these issues. This was demonstrated through a novel, cloud-based design of a core that remains complaint with the standard radio access network (RAN) interfaces. However, due to the novelty of this work, the question remains as to whether this radically different paradigm can be successfully adopted by network operators.Key questions that remain unanswered include whether moving to the cloud will impact the security of the core, whether the multi-vendor approach pioneered with the Open-RAN initiative can be adapted to a cloud-based core, and whether the resilience and upgradable of the core can be improved upon in a cloud environment. I will answer these questions by studying the performance, security, resilience and upgradability of the core design in real operational environments, focusing on the perspective of the mobile network operator. This study will result in both modifications to the novel core design, and perspectives beneficial for other critical infrastructure that is deployed in cloud environments.The second pillar of my work expands the focus beyond the core and into the RAN. Existing mobile devices have limited ability to seamlessly switch between different networks. This includes networks using the same radio access technology (RAT), such as moving between a public and a private 5G network, and multi-RAT networks (e.g: moving from 5G to WiFi). The increasing proliferation of RATs (WiFi, 4G, 5G, light-based Li-Fi, and new non-terrestial technologies such as Starlink and OneWeb) has resulted in the need for a new approach to managing seamless roaming between different networks. Key questions include how to adopt seamless roaming from a standards-based approach, where is the optimal part of the network to modify to achieve this (on the mobile network operator side, or on the user side, or both), and what additional technologies are needed to ensure this vision. I will answer this from a mobile network operator perspective, focusing, as with my work on the mobile network core, on the operational aspects including the deployment and management of potential solutions. This research will be achieved through (a) a study of existing roaming technologies and (b) the design and evaluation of new roaming technologies to address the limitations of existing solutions. A planned campus-scale private 5G network, integrating with an existing campus-scale WiFi deployment, will assist in the study and evaluation of multi-RAT networks while single-RAT networks will be studied through roaming between public 5G networks and the private 5G deployment.The final pillar of my work focuses on developing tools and infrastructure for research that will be used to achieve the first two pillars. Software tools include frameworks necessary to develop, test and evaluate core networks, and tools to emulate real-world networks which are necessary for simulations of roaming scenarios. Physical infrastructure includes lab-scale and campus-scale mobile networks. Key questions include: what tools are required for research into mobile core networks and roaming, how should physical infrastructure be setup to best enable this research, and how should the tools and knowledge gained from the infrastructure be disseminated to best enable future research, additional studies and reproducibility