
KDDI, Samsung AI Boosts 5G Network Speed by Up to 52% in Tokyo Trial
KDDI has confirmed that Samsung's AI RAN Speed Optimizer significantly boosted 5G Standalone network performance during peak traffic hours, based on live commercial network tracking in Tokyo. The two companies completed a multi-month field trial applying automated intelligence to individual base stations instead of traditional cluster-wide configurations.
Kazuhiro Furuhata, Chief Network Officer at KDDI, said the trial proved that individual cell tuning — a long-standing industry challenge — is now achievable through AI. "Combining KDDI's accumulated expertise in network innovation and Samsung's technical leadership, this field trial proves that individual tuning for cells has now become a reality through the integration of AI," he said, adding that KDDI remains committed to advancing AI-based technologies for its customers.
The trial covered dense urban financial districts, residential suburbs, and peripheral rural zones, with continuous telemetry collected to train predictive models against real-world congestion and physical obstructions.
Peak throughput results
Metric | Result |
|---|---|
Average downlink throughput increase (trial-wide) | 31% |
Maximum downlink throughput increase (dense urban areas) | 52% |
Spectrum used | 100 MHz, mid-band 3.7 GHz TDD |
Cells monitored | Hundreds |
The biggest gains came in high-density urban sectors with severe signal attenuation and complex multipath reflections.
How the technology works
The AI RAN Speed Optimizer runs within the Samsung CognitiV Network Operations Suite. It ingests live site data, calculates the optimal radio frequency settings for each location, and applies updates automatically — without manual script editing or field technician visits.
This replaces traditional optimisation cycles that typically require iterative drive testing and manual cluster tuning, instead matching power levels, beamforming angles, and scheduling priorities to real-time demand.
KDDI's existing fully virtualised RAN architecture, which separates software from proprietary baseband hardware, gave the trial the computational agility needed to push real-time changes across hundreds of sites at once.
June Moon, Executive Vice President and Head of R&D, Networks Business at Samsung, said: "Since 2024, we have been actively testing and training our AI-powered RSO technology in the field and we have demonstrated breakthrough progress in its capabilities. This trial with KDDI exemplified how Samsung's AI-powered innovation can bring advanced optimisation to live commercial networks."
What's next
Samsung and KDDI plan to expand the automated optimisation models to broader commercial bands and multi-carrier aggregation scenarios, as operators look to control costs while maximising capacity from existing spectrum amid growing 5G Standalone demand.
[Source: telecomstechnews]
Topics in this story
Article Details
Category: Telecom
Published: 2 July 2026
Time: 12:43 pm
Author: Usama Haider
More Stories



