RSaaS: Radio Signal as a microService

RSaaS：无线电信号作为微服务

• 批准号: 82300
• 金额: $ 12.63万
• 依托单位: RANPLAN WIRELESS NETWORK DESIGN LTD
• 依托单位国家: 英国
• 项目类别: Collaborative R&D
• 财政年份: 2020
• 资助国家: 英国
• 起止时间: 2020 至 无数据
• 项目状态: 已结题
• 来源: https://gtr.ukri.org/projects?ref=82300
• 关键词: RSaaS; Radio; Signal; microService

COVID-19 has had a profound impact on mobile network operations: 1) network loads have surged far beyond capacity; 2) network energy consumption has increased, resulting in a higher CO2 emission; and 3) engineers have limited access to sites, in particular, for in-building wireless networks, which has caused significant problems on network deployment and maintenance.The above points emphasise that: 1) mobile networks must be able to adapt to changes; 2) the planning, commissioning, and operations of radio access networks, need to be automated so that the number of physical site visits, can be minimised.To this end, many what-if scenarios need to be carefully evaluated before changes of network parameters are commissioned, e.g., how the changes of network parameters will impact the network coverage. Radio propagation models play a central role in evaluating these what-if scenarios. Furthermore, network automation calls for interactions between radio signal prediction with an eco-system of data analytics, optimisation, and operations support system (OSS) software tools, which makes a migration to a cloud computing platform based on microservices, imperative.In this project, we will enhance Ranplan's **world first** combined indoor-outdoor radio propagation engine with machine learning algorithms and implement the new radio propagation engine as a **microservice** in a cloud-computing platform. The first part of this innovative approach will lead to a universal radio signal (or interference) prediction engine that works in **all scenarios** (indoor-outdoor, millimetre waves, etc), while the second part will make it **universally available** so that it can easily interact with other services and make use of the vast computing power on the cloud. develop a stochastic radio propagation model based onMobile networks are a key enabler for economy, social connectivity, remote working/ teaching. As the use of network increase, so does energy consumption. Recent studies suggest that mobile networks will produce 320 million Tonnes of CO2 by the end of the current decade. Ranplan believes that the proposed project forms a foundation to enable network operation automation, resulting in a substantial reduction of use of equipment and CO2 emissions.

COVID-19对移动的网络运营产生了深远的影响：1）网络负载激增，远远超出容量; 2）网络能耗增加，导致二氧化碳排放量增加; 3）工程师进入站点的机会有限，特别是建筑物内的无线网络，这对网络部署和维护造成了重大问题。上述几点强调：1）移动的网络必须能够适应变化; 2）无线电接入网络的规划、调试和操作需要自动化，使得物理站点访问的数量可以最小化。为此，在调试网络参数的变化之前，需要仔细评估许多假设场景，例如，网络参数的变化将如何影响网络覆盖。无线电传播模型在评估这些假设场景中发挥着核心作用。此外，网络自动化要求无线电信号预测与数据分析、优化和运营支持系统（OSS）软件工具的生态系统之间进行交互，这使得迁移到基于微服务的云计算平台势在必行。在该项目中，我们将加强兰普兰的 ** 世界第一 ** 结合室内-室外无线电传播引擎与机器学习算法，并实现新的无线电传播引擎作为一个 ** 微服务 ** 在云计算平台。这种创新方法的第一部分将导致一个通用的无线电信号（或干扰）预测引擎，该引擎适用于 ** 所有场景 **（室内-室外，毫米波等），而第二部分将使其 ** 普遍可用 **，以便它可以轻松地与其他服务进行交互，并利用云上的巨大计算能力。移动网络是经济、社会连接、远程工作/教学的关键推动者。随着网络使用的增加，能源消耗也在增加。最近的研究表明，到本十年末，移动的网络将产生3.2亿吨二氧化碳。Ranplan认为，拟议的项目为实现网络运营自动化奠定了基础，从而大幅减少了设备的使用和二氧化碳排放。