AI-Powered Ericsson Uplink Optimization Trial Delivers Performance Improvements Across KDDI 4G and 5G Networks

AI-Powered Ericsson Uplink Optimization Trial Delivers Performance Improvements Across KDDI 4G and 5G Networks

(IN BRIEF) Ericsson and KDDI have successfully completed a large-scale AI-driven uplink optimization field trial across KDDI’s live 4G and 5G commercial network in Japan. Using Ericsson’s Uplink Interference Optimizer rApp on the Ericsson Intelligent Automation Platform, the companies achieved measurable improvements in uplink throughput, spectral efficiency, and 5G signal quality under real-world traffic conditions. The trial covered around 2,800 network cells and demonstrated the readiness of AI-powered optimization tools to support higher levels of autonomous network operations. The deployment also validated support for third-party rApps, including the Japan-developed FYRA Suite, reinforcing KDDI’s commitment to open and interoperable O-RAN ecosystems.

(PRESS RELEASE) STOCKHOLM, 11-May-2026 — /EuropaWire/ — Ericsson and KDDI have completed a large-scale field trial in Japan demonstrating the effectiveness of AI-driven uplink optimization across commercial 4G and 5G networks, marking a significant step toward higher levels of autonomous network operations.

The trial centered on Ericsson’s AI-powered Uplink Interference Optimizer (UIO) rApp operating on the Ericsson Intelligent Automation Platform (EIAP). Conducted during the first quarter of 2026, the deployment covered approximately 1,500 5G cells and 1,300 4G cells across multiple technologies and frequency bands within KDDI’s live commercial network environment.

The project demonstrated that AI-based uplink interference management can improve network performance under real-world traffic conditions while maintaining operational stability. Ericsson stated that the AI model continued to perform effectively despite uplink traffic levels increasing by around 10 percent during the testing period, highlighting the robustness of the system under sustained load.

According to the trial results, the UIO rApp delivered measurable improvements in network throughput and spectral efficiency across both mobile generations. Average uplink throughput gains reached 9.6 percent in 4G environments and 3.1 percent in 5G networks, contributing to improved end-user connectivity and data performance. The companies also reported a 27 percent improvement in 5G Signal-to-Interference-plus-Noise Ratio (SINR), alongside stronger uplink modulation and spectral efficiency.

The trial marked several industry milestones. Ericsson said it was the first global deployment of the AI-driven UIO rApp within a live commercial network, as well as the first use of the Ericsson Intelligent Automation Platform in a live operational environment in Japan. The deployment also demonstrated how AI-driven optimization tools can support communication service providers pursuing advanced levels of network automation.

Using an adapted version of the TM Forum Autonomous Network Levels Evaluation Tool (ANLET) methodology, the UIO rApp was assessed as reaching Level 3.86 autonomy for uplink optimization tasks. Ericsson believes this positions the technology as an important stepping stone toward Autonomous Networks Level 4 operations, where network management functions become increasingly self-optimizing and self-operating.

Jean-Christophe Laneri said the field trial confirmed the operational readiness of the UIO rApp and EIAP platform for future autonomous network deployments. He noted that the technology introduces more advanced and reliable automated optimization capabilities aimed at improving subscriber experience while supporting next-generation network automation strategies. Ericsson also confirmed that the Uplink Interference Optimizer rApp has now reached General Availability status for global deployment, with additional AI-driven rApps expected to follow.

Hiroaki Ando highlighted the growing importance of highly reliable and low-latency uplink connectivity for emerging Physical AI applications, which have become an increasingly important focus area within Japan’s broader semiconductor and digital industry initiatives. He said Ericsson intends to continue collaborating closely with KDDI on future network innovation projects.

Another significant outcome of the field trial was the successful integration of a third-party rApp developed by Japan-based FYRA. The deployment validated the Ericsson Intelligent Automation Platform’s compliance with the R1 standard and demonstrated support for both Ericsson-developed and third-party applications within an open ecosystem framework.

Alongside Ericsson’s Uplink Interference Optimizer, KDDI tested the FYRA Suite rApp, reinforcing the operator’s strategy of supporting an interoperable, O-RAN-aligned radio access network ecosystem. Ericsson noted that the combination of internal and external rApps highlights how open automation platforms can encourage broader innovation by enabling regional technology developers to create scalable applications for global markets.

The successful field trial further strengthens the role of AI-driven automation in the evolution of mobile networks as operators increasingly seek more efficient, intelligent, and self-managing infrastructure capable of supporting future digital services and high-performance connectivity demands.

MORE INFORMATION AT:
Ericsson Newsroom
media.relations@ericsson.com  (+46 10 719 69 92)
investor.relations@ericsson.com  (+46 10 719 00 00)

ABOUT ERICSSON:
Ericsson’s high-performing, programmable networks provide connectivity for billions of people every day. For 150 years, we’ve been pioneers in creating technology for communication. We offer mobile communication and connectivity solutions for service providers and enterprises. Together with our customers and partners, we make the digital world of tomorrow a reality.
www.ericsson.com

SOURCE: Telefonaktiebolaget LM Ericsson