Connect with us

Company News

Rakuten demonstrate 25% energy savings through AI model

Rakuten Mobile and Rakuten Symphony today announced the successful demonstration of an Artificial Intelligence and Machine Learning model that enables energy savings of up to 25% via an interface to RAN Intelligent Controller (RIC) that configures antenna use.

In a pioneering initiative for the industry, Rakuten Symphony’s implemented RAN Intelligent Controller (RIC) is now able to set a policy via an intelligent interface that allows communication between rApps and xApps, providing insights on the usage of specific cell sites, allowing the reduced usage of specific elements and consequently making significant energy savings.

The organizations demonstrated results of the trial – as part of a project commissioned by the Japan National Institute of Information and Communications Technology (NICT)*1 – at MWC Barcelona 2024, the largest gathering of mobile connectivity operators, vendors, and research and academic institutions worldwide.

Sharad Sriwastawa, Co-CEO, Rakuten Mobile, and President, Rakuten Symphony, said, “We would like to thank the Japan National Institute of Information and Communications Technology for their commitment to innovation in telecom and to pushing the boundaries of research in automation and Open RAN. From day one of their existence, Rakuten Mobile and Rakuten Symphony have been committed to building automated, intelligent and open networks that are powered by software and run in the cloud. This research reinforces the commitment of both organizations to sustainable, automated networks that can be made more sophisticated through artificial intelligence.”

rApps and xApps are used to control the radio access base stations to quickly change relevant parameters. xApps – software delivering response times of sub-one-second, running on a near-real time RIC – used in the trial have been able to monitor a cell’s traffic in near-real-time. Through machine learning models, they can intelligently interpret historical traffic patterns to modify the configuration of antennas without impacting the subscriber experience, saving energy in the process. rApps run on a non-Real Time RAN Intelligent Controller, with control loops on a time scale of one second and longer, making updates that do not require immediate responses.

The organizations have developed a machine learning model which has been trained to interpret network traffic patterns and subsequently determine energy-saving policies. These policies are communicated via an intelligent interface from an rApp to an xApp*2. The xApp subsequently adjusts antenna configurations to conserve energy, leading to enhanced Open RAN performance and more efficient, energy-saving network operations.

The demonstration proves that applications on the RIC Platform can effectively manage RAN efficiency and reduce network power consumption, benefiting both 5G and 4G Open RAN networks. Power savings of almost 25% have been observed when the feature is activated under optimal conditions.

Rakuten Mobile and Rakuten Symphony began collaborating on the RIC Platform and Applications development in February 2023, and the project has been a significant step towards advancing wireless communication technologies, specifically in the area of Open RAN.

Rakuten’s RIC platform includes both Non-Real-Time (Non-RT RIC) and Near-Real-Time (Near-RT RIC) components, developed in accordance with O-RAN ALLIANCE specifications. The AI/ML model for the rApp is defined and trained using Rakuten Symphony’s AI Platform product. Rakuten’s RIC platform supports Non-RT-RIC and Near-RT-RIC capabilities independently. It integrates seamlessly with 5G as well as 4G Open RAN, is compliant with O-RAN specification and interfaces (can work with any O-RAN compliant O-CU/O-DU), provides seamless support for onboarding RIC applications (rApps and xApps) and is a highly scalable platform that employs a micro-services/container-based architecture with flexible deployment options.

CT Bureau

Click to comment

You must be logged in to post a comment Login

Leave a Reply

Copyright © 2024 Communications Today

error: Content is protected !!