[DigitalToday reporter Jin-ho Lee] A machine-learning-based methodology developed by South Korean researchers to assess and predict frequency availability has been adopted as a new reference methodology in international radio spectrum management.

The Electronics and Telecommunications Research Institute (ETRI) said on Tuesday that a new report, "Spectrum Availability Assessment and Prediction Methodology", was approved at an International Telecommunication Union Radiocommunication Sector (ITU-R) Spectrum Management Study Group (SG1) meeting held in Geneva, Switzerland, from June 3 to 11.

Spectrum availability refers to the degree to which frequencies can be used for wireless communications in a specific region and time period. As next-generation wireless services such as AI services, 6G mobile communications and low-Earth-orbit satellite communications expand, the importance of assessment and forecasting technology is growing to use limited frequencies efficiently.

ETRI has researched technology to assess and predict spectrum availability using actual mobile communications frequency usage data. It developed technology to assess the balance between frequency demand and supply in LTE networks, as well as machine-learning-based LTE frequency usage forecasting technology, and published the results in international academic journals.

Based on those research results, ETRI prepared a spectrum availability assessment and prediction methodology that can be used internationally. The methodology was reflected as core content in the new ITU-R report.

The report was drafted based on ETRI's machine-learning-based spectrum availability assessment and forecasting technology published in international academic journals. The report includes methods for assessing and predicting the degree to which frequencies can be used in specific regions and time periods using machine learning, and will be issued as an official report after ITU-R's final editing process.

The research was carried out with support from the Ministry of Science and ICT's project to develop technologies to secure and supply spectrum. ETRI plans to continue related technology development and international standardisation activities, including AI-based mobile traffic pattern forecasting, digital virtual-cell-based spectrum efficiency forecasting, and network capacity analysis by generation of mobile communications technology.

Park Seung-geun (박승근), head of ETRI's Radio Research Division, said the ITU-R report approval was an achievement showing that ETRI's data-based spectrum management technology has been recognised internationally. He said ETRI will advance intelligent spectrum management technology so that the frequencies needed for wireless AI communications can be supplied in a timely way and in appropriate amounts.