Fujitsu, which has a successful track record of introducing radio unit (RU) products to the global market, will incorporate this technology into its O-RAN-compatible Service Management and Orchestration (SMO) system, FUJITSU Network Virtuora Service Management and Orchestration. The Japanese multinational will begin a phased global rollout of the technology to mobile network operators starting in November 2024, leveraging its existing RU footprint.


This application consists of three key technologies:

An AI-based technology that can estimate the quality of experience (QoE) of smartphone users on your networks in real time.
A technology that detects early signs of increased communication traffic and proactively prevents network quality degradation.
A technology that detects network quality degradation and redefines the area covered by edge base stations to maintain optimal quality.

Fujitsu verified the effectiveness of these technologies in August 2024 using real commercial data from mobile network operators (4) under conditions very similar to real operating environments.

This application will ensure a smooth and reliable connection for mobile network users, not only during normal operations but also during emergencies and periods of peak network traffic, ultimately improving user convenience, satisfaction, and safety in critical situations. For mobile network operators, the application will reduce operating costs and save energy through optimized operations.


Background

As digital transformation (DX) accelerates globally, the adoption of 5G mobile networks is rapidly expanding as a key infrastructure component. Mobile network operators are expected to further enhance their capabilities, including ultra-low latency and simultaneous connectivity, while ensuring network quality at both the user and application levels. Furthermore, the Radio Access Network (RAN) domain is undergoing an open and virtualized transformation based on the O-RAN concept, leading to anticipated reductions in total cost of ownership (TCO).

The three technologies in this application operate in the RIC (5) deployed within the O-RAN compliant SMO, contributing to intelligent automation and RAN self-sufficiency.

Development achievements


1. Real-time QoE estimation and quality assurance using AI

This technology estimates QoE in real time and automatically moves users to other areas of the base station network when QoE degradation is detected. This world first enables the creation of AI models capable of easily estimating QoE for individual applications by selecting feature values ​​from statistical data (KPIs) calculated from high-speed packet analysis for 100 Gbps RAN traffic. This approach allows for flexible adaptation to diverse applications.


By accurately understanding each user's QoE and allocating the necessary resources, this technology ensures user comfort and satisfaction while eliminating resource over-allocation. The result is a 19% increase in the number of users that can be hosted per base station (6).

figure-1-fujitsu-w

2. Proactive activation/deactivation of base stations to maintain quality and save energy
Fujitsu has developed a technology that uses AI to anticipate increased communication traffic and proactively activate previously inactive base stations to prevent degradation of user communication quality.

Previously, energy savings were achieved by monitoring traffic in each zone in real time and putting unnecessary base stations into standby mode. This world-first technology goes a step further by detecting unusual increases in pedestrian traffic, such as those associated with local events, and predicting subsequent traffic increases at the network level (7). This predictive technology has been shown to successfully activate base stations in advance without impacting user quality 99.8% of the time during the verification period.


This allows for the precise activation and deactivation of base stations based on traffic conditions, achieving a balance between maintaining QoE and saving energy when combined with the energy-saving application announced by Fujitsu in December 2023 (8).


3. Detecting Service Quality Degradation and Redesigning Zones for Service Quality Maintenance.
Traditional single-cell fault detection technologies (9) struggled to differentiate between simple load reductions and actual faults. Fujitsu's new technology addresses this problem by comparing traffic trends in surrounding cells using AI, achieving a fault detection accuracy rate of over 92%. This technology supports both supervised learning with limited fault data and unsupervised learning. By understanding the service impact, including cell overlap, it is possible to determine which zones should be restored first.

When this anomaly detection technology identifies areas experiencing a significant service impact, it uses a radio propagation prediction model that considers path loss(10) in the actual field, as well as the direction and load conditions of surrounding cells, to calculate the optimal tilt angle for those cells. This minimizes the impact on service quality caused by faulty cells. As a result, the recovery time from anomalies such as equipment failure, which previously took 24 hours, has been reduced to less than one hour, minimizing the impact on users.

figure-2-fujitsu-w

 


Future plans:
Fujitsu will continue to contribute to a safer and more sustainable society by providing O-RAN products that utilize AI-centric technologies for networks that support applications and services across all sectors



Grades

[1] NEDO-led Project: Project Name: Research and Development Project for Enhanced Infrastructures for Post-5G Information and Communication Systems (JPNP20017) (Commissioned) Period: FY2021 - 2024 Overview: Research and Development Project for Enhanced Infrastructures for Post-5G Information and Communication Systems
[2] O-RAN: O-RAN refers to a radio access network (RAN) built on the basis of open interface specifications developed by the O-RAN Alliance.
[3] Service Management and Orchestration (SMO): In the O-RAN architecture, Service Management and Orchestration (SMO) defines a system that manages the operation, maintenance, and optimization of the RAN, as well as its lifecycle.
[4] Commercial Mobile Network Operator Data: This refers to data collected by mobile network operators during their actual operations. This data is used in the AI ​​development for this research and development project.
[5] RIC: Abbreviation for RAN Intelligent Controller, a controller that autonomously executes and controls RAN parameter settings and operational optimization.
[6] A 19% increase in the number of users that can be hosted per base station: This figure was calculated from a comparison of cell-based performance measurements and QoE-based performance measurements in Fujitsu's lab environment.
[7] Grid: A grid is a section of a map based on latitude and longitude.
[8] The power-saving application announced by Fujitsu in December 2023: This refers to a power-saving application that uses Fujitsu's AI technology to estimate communication traffic based on the distribution of users' location information across the mobile network. Press release: "Fujitsu leverages AI technology to make power savings a reality in network operations."
[9] Cell: A cell is the smallest unit of a communication area formed by radio waves from a base station.
[10] Path Loss: Path loss refers to the attenuation of radio waves as they travel through the air or are blocked by obstacles.