Samsung, KDDI complete AI-based 5G SA network optimisation trial

Declaring that the fruits of their work have demonstrated how artificial intelligence (AI) products can deliver benefits in daily operations and actualise a transformative leap in enhancing network performance, Samsung Electronics has announced the successful trial of its AI-powered RAN Speed Optimiser (RSO) on the commercial 5G Standalone (SA) network of KDDI in Japan.

Conducted in and around Tokyo, the joint trial began in late 2025, lasting for several months, spanning dense urban, suburban and rural areas. It was also based on 100MHz of 3.7 GHz TDD spectrum across hundreds of cells.

Samsung’s AI-powered RSO is built to enhance network performance through what is described as a distinctive parameter optimisation per cell. This is claimed to offer a significant change and advancement, where AI-tuned customisation moves beyond the conventional cluster-level approach that applies the same parameter settings across multiple cells.

By contrast, Samsung’s RSO uses its own AI-based prediction model to analyse site environment data automatically and recommend optimised parameters tailored to each individual cell. The RSO is part of the Samsung CognitiV Network Operations Suite (NOS), a set of diverse AI-powered network automation offerings, AI agents and intelligent applications.

Talking about the tests and its results, Samsung said the diverse traffic and network environments allowed robust AI model training to be carried out across real-world conditions. During peak hours, Samsung’s RSO is said to have delivered an average increase of 31% in 5G downlink throughput across the overall trial area, with gains reaching a maximum increase of 52% in dense urban areas.

In addition to what they called “substantial” performance gains, the partners insisted that the trial highlighted the potential and effectiveness of AI-based services to enable more precise and adaptive network optimisation in commercial environments.

The partners said that going forward, they can benefit operators by automatically adjusting to changing conditions across different network scenarios, allowing for reduced manual intervention and greater cost efficiency. For users, this means faster and more reliable connections, whether they are streaming videos, browsing the web or making calls.

Building on the outcomes delivered in the trial, Samsung and KDDI plan to continue evaluating AI-based optimisation technologies for broader commercial network applications. The companies’ collaboration on fully virtualised network deployments provides KDDI with an optimal foundation for advancing towards AI-native network operations.

“Combining KDDI’s accumulated expertise in network innovation and Samsung’s technical leadership, this field trial proves that individual tuning for cells – a long-standing industry challenge – has now become a reality through the integration of AI,” said Kazuhiro Furuhata, chief network officer at KDDI. “Moving forward, we remain committed to pushing the boundaries of AI-based technologies to continuously elevate network experience for our customers.”

June Moon, executive vice-president and head of R&D for networks business at Samsung Electronics, said: “Samsung continues to help operators like KDDI build intelligent and efficient networks by weaving in AI-powered innovation.”

“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 while facilitating steady, seamless connectivity in an array of different network environments.”

The Samsung Networks 5G portfolio encompasses purpose-built RAN, vRAN, Open RAN, AI-RAN and cloud-native core to private network services and AI-powered automation tools.