KT said on February 23 that it has successfully verified an artificial intelligence-based beam pattern optimization technology in collaboration with global telecommunications equipment provider Nokia.
The technology applies AI to precisely reflect regional and environmental characteristics in optimizing base station coverage, an approach viewed as a key driver for enhancing wireless quality.
The newly validated solution uses AI to analyze real-time traffic flows and diverse network data, enabling dynamic reconfiguration of beam patterns. Beam patterns determine the direction, width and strength of signals transmitted by base station antennas, making them a critical factor in network performance.
Massive multiple-input multiple-output (mMIMO) technology, widely used in 5G base stations, leverages multiple transmit and receive antennas to simultaneously deliver various beams to numerous users, significantly improving capacity and efficiency. However, with tens of thousands of possible beam pattern combinations available within a single cell, identifying the optimal configuration remains highly complex.
The AI-based technology verified by KT and Nokia employs reinforcement learning to efficiently explore vast beam pattern candidates and select optimal outcomes. Through periodic learning, the system autonomously adjusts beam strategies in response to changes such as time-of-day traffic fluctuations or special events. This enables improved wireless signal quality, more stable coverage and enhanced capacity performance within the same hardware environment.
The verification, conducted in February, involved close collaboration between KT and Nokia’s global research organizations to assess the technology’s effectiveness and stability. KT expects the solution to contribute significantly to its AI-RAN advancement strategy by improving wireless quality and boosting network operational efficiency.
The two companies plan to conduct field validation on commercial networks in the second half of the year and expand cooperation to strengthen core AI-RAN technologies in preparation for the 6G era.
Lee Jong-sik, Executive Vice President and Head of KT’s Future Network Research Institute, said the verification demonstrated that AI can configure beam patterns more precisely in response to network conditions. He added that KT will continue to advance AI-RAN technologies with Nokia and expand commercial applications to deliver tangible improvements in wireless service quality for customers.
