NTT DOCOMO, INC. and Samsung Electronics have introduced the profitable validation of AI-based radio entry community (AI-RAN) optimization expertise on the person stage, marking a major development in community autonomy. This improvement is poised to allow networks to autonomously acknowledge person contexts and repair necessities, delivering optimum communication environments.
The validation represents a vital step towards predictive community operations, as AI-driven community optimization turns into extra prevalent within the upcoming 6G community period. Traditionally, person units linked to a single cell had been subjected to uniform community configurations, typically resulting in service high quality points corresponding to connection drops in areas with weak indicators. The new AI-RAN optimization expertise addresses these challenges by working on the individual-user stage, mechanically making use of configurations tailor-made to real-time wi-fi circumstances and repair calls for.
The AI system learns every person’s motion patterns and repair utilization habits, predicting potential service high quality degradation earlier than it happens. In a validation performed in Japan in January 2026, utilizing simulations based mostly on knowledge from DOCOMO’s industrial community and an area 5G trial subject, the businesses lowered the frequency of communication-speed degradation from 13.1% to 7.2%, an enchancment of 5.9 proportion factors in comparison with operations with out the expertise.
By using aggregated knowledge from person units, the system analyzes particular person person contexts and motion tendencies to detect early indicators of throughput degradation earlier than service high quality declines. For occasion, if a person’s transmission pace is predicted to fall under the required threshold, the system will mechanically change to the optimum configuration, corresponding to a frequency band, to fulfill service calls for.
Unlike conventional strategies that required bulk knowledge assortment and processing, the brand new process selectively gathers solely vital info based mostly on particular person points, decreasing community pressure whereas acquiring vital knowledge wanted for AI-driven throughput degradation prediction and community optimization. In February, the businesses contributed to 3GPP discussions on knowledge assortment strategies, advancing 6G expertise standardization efforts.
Moving ahead, DOCOMO and Samsung plan to proceed advancing this expertise and exploring future 6G use instances. The two firms additionally intend to contribute the AI-based throughput degradation prediction expertise and environment friendly knowledge assortment applied sciences to 3GPP discussions, aiming to play a number one position in realizing AI-centric networks the place networks autonomously study, predict, and optimize their habits via AI.
Masafumi Masuda, Senior Vice President and General Manager of the 6G-Tech Department at NTT DOCOMO, highlighted the achievement as a major step towards extra adaptive and user-centric communication experiences. JinGuk Jeong, Executive Vice President and Head of the Advanced Communications Research Center at Samsung Electronics, emphasised the potential for enhanced perceived high quality on the person stage via AI-driven particular person person concerns.