Capgemini Engineering and Intel Architecture have launched the industry’s first machine learning-based RAN application to improve the efficiency of the 5G spectrum. Unveiled at the Mobile World Congress (MWC), the application, named “Project Marconi” aims to give mobile operators “a significant advantage” in monetising 5G services faster, the companies claim.
Capgemini says that the solution meets O-RAN (Open Radio Access Network) guidelines to maximise 5G spectrum efficiency. It also boosts subscriber quality of experience (QoE) with real-time analytics. With this new application, the efficiency of the 5G spectrum is expected to increase by 15%.
Walid Negm, Chief Research and Innovation Officer at Capgemini Engineering, said: “Our teams worked closely with Intel to create a truly innovative solution that can really move the needle for operators. We gathered and utilised over one terabyte of data and conducted countless test runs with NetAnticipate5G to fine-tune the predictive analytics to meet diverse operator requirements. In short, machine learning can be deployed for intelligent decision-making on the RAN without any additional hardware requirements.
“This makes it cost-efficient in the short run and future proof in the long run as we move into Cloud Native RAN implementations”, he said.
Project Marconi is the first artificial intelligence and machine learning (AI/ML) based radio network application for the 5G Medium Access Control (MAC) scheduler and is optimised with Intel AI Software and 3rd Generation Intel Xeon Scalable processors.
Network providers are increasingly looking for solutions to develop and gain 5G services faster and more easily, investing heavily in the 5G spectrum to be able to do so. According to the Global Mobile Suppliers Association, the total value of spectrum auctions reached over US$27bn last year. Capgemini says that its Project Marconi solution in partnership with Intel Architecture aims to increase the amount of traffic handled by each individual cell. It also allows operators to “serve more subscribers and deliver an outstanding experience” while launching new Industry 4.0 services such as Ultra Reliable Low Latency Communications (URLLC) and enhanced Mobile Broadband (eMBB) use cases.
Capgemini deployed its NetAnticipate5G and RATIO O-RAN platform to introduce advanced AI/ML techniques. The AI-powered predictive analytical solution aims to forecast and assign the appropriate modulation and coding scheme (MCS) values for signal transmission through forecasting the user signal quality and mobility patterns accurately. “In this way, the RAN can intelligently schedule MAC resources to achieve up to 40% more accurate MCS prediction and yield to 15% better spectrum efficiency in the case studies and testing”, Capgemini said. This results in delivering faster data speeds, better and more consistent QoE to subscribers, and robust coverage for use cases that rely on low latency connectivity, such as robotics-based manufacturing and V2X (vehicle-to-everything).
Cristina Rodriguez, VP of Wireless Access Network Division at Intel, said: “Our 3rd Gen Intel Xeon Scalable processors with built-in AI acceleration provide high performance for deep learning on the Net Anticipate 5G platform. Together, our collaboration delivered ultra-fast inference data to enhance the Open-Source ML libraries resulting in an intelligent RAN that can predict and quickly react to subscriber coverage requirements while reducing TCO”.