KT announced that it has started verifying "Artificial Intelligence Radio Access Network" (AI-RAN) technology on its commercial 5G network. This marks the first instance in the country of implementing AI-RAN on a commercial 5G network.
AI-RAN is next-generation communication network technology that analyzes traffic data flowing between base stations and network servers in real time to provide optimal network environments. Base stations equipped with AI-RAN technology optimize wireless resources by using dedicated AI engines to monitor radio conditions and traffic patterns in real time.
KT plans to conduct this technology verification in the Naju area of South Jeolla Province using the dedicated AI processors built into Nokia's latest 5G base station equipment. Based on the verification results, KT aims to launch AI-RAN commercial services and gradually expand the areas of application.
AI-RAN is also considered an essential technology for the 6G era. This 5G-based AI-RAN verification is a necessary preliminary step for that purpose. KT intends to optimize and upgrade AI-RAN technology on its commercial network and proactively establish the technical specifications and operational standards required for 6G.
KT will test key AI-RAN functionalities, including ▲machine learning-based channel estimation ▲machine learning-based MU-MIMO operation on its commercial network. Machine learning-based channel estimation involves learning complex radio patterns to more precisely identify variations in signal strength according to the wireless environment. The quality of wireless communication depends on how accurately the process of the base station's signals reaching the device is predicted and transmitted. Previously, this was analyzed using mathematical calculations, but there was a drawback as errors could increase in areas where signals scattered complexly due to obstacles like mountains or high-rise buildings. Solving this with AI models can reduce data loss and delay, making stable communication quality possible. This allows for transmitting more data on the same frequency.
MU-MIMO (Multi-User Multiple Input Multiple Output) is technology that sends data to multiple users simultaneously using a single frequency and time resource. Machine learning-based MU-MIMO uses AI to instantly analyze the signal direction and traffic situation of each device to find combinations of users with minimal interference. This increases the amount of data the base station can deliver within the same time frame, expanding overall processing capacity and allowing users to experience faster speeds and superior quality.
Kwon Hye-jin, head of KT's Network Strategy Department, said, "AI RAN is essential technology that enhances customer-perceived quality while preparing for the 6G era, and KT will be the first to verify and adopt AI RAN on its commercial network, leading network innovation and continuously strengthening its technological leadership."
Source: https://biz.chosun.com/en/en-it/2025/08/26/VOACVNANXNFDPOXVK5GACKIHQI/
