KT successfully demonstrated next-generation wireless transmission and reception technology using AI, proving in a commercial environment that AI-based wireless networks can improve actual communication performance.

KT announced on the 11th that it had completed a demonstration of AI-based wireless transmission and reception technology in collaboration with global wireless communications company Qualcomm Technologies and German telecommunications equipment manufacturer Rohde & Schwarz. This demonstration applied AI-based channel state information (CSI) compression technology, which is being promoted by the international standardization organization 3GPP in Release 20.

This technology uses AI to learn the complex wireless environment between base stations and terminals, thereby improving the accuracy of CSI transmitted from terminals to base stations. This enables base stations to perform more precise transmission control, ultimately improving downlink performance.

For the demonstration, KT and Qualcomm built a test environment that linked a next-generation mobile platform with built-in AI capabilities and Rohde & Schwarz's AI base station test system. KT applied an AI model reflecting its 5G network operation experience and base station configuration information.

As a result, the accuracy of wireless channel status information was improved, and downlink performance was confirmed to have improved by approximately 50%. This is significant in that it verifies AI-RAN technology based on 3GPP standards under actual network conditions.

KT plans to capitalize on this achievement by strengthening global cooperation for 6G standardization and the development of AI-based wireless technologies. Furthermore, the company plans to gradually evolve into an autonomous network powered by AI.