□ The DGIST (President Lee Kun-woo) Division of Mobility Technology research team, led by Kim Sang-dong and Kim Bong-seok, and the Republic of Korea Naval Academy’s team, headed by Professor Choi Yeong-doo, jointly developed the ‘Extrapolation-based Doppler Resolution Enhancing Algorithm’ for FMCW radar, achieving higher performance than existing ultra-high resolution technology.
□ This research introduces a technology that improves the detection accuracy of radars without requiring additional complex computation or equipment. It aims to enhance the performance of radar systems used in various intelligent mobility applications, such as unmanned aerial vehicles, unmanned ships, and autonomous vehicles.
□ Existing radars analyze the ‘Doppler effect’ to determine a target's speed, but the FFT-based approach has resolution limitations in speed discrimination. To address this issue, the DGIST-Naval Academy joint research team proposed a new algorithm that can enhance Doppler resolution without increasing observation time by using signal extrapolation techniques.
□ As a result, the proposed technology successfully reduced the root mean square error (RMSE) by up to 33% and lowered the target undetected rate by as much as 68% compared to the current approach. In particular, it maintained a computational complexity similar to that of the existing FFT method, simultaneously ensuring fast processing speeds and high efficiency.
□ This technology can effectively solve the problem of signal overlap between targets moving at similar speeds, especially when unmanned aerial vehicles or radar systems detect multiple objects at the same time. With this capability, it can significantly improve the ability to distinguish closely located targets and enhance recognition accuracy, marking a new milestone in the development of high-resolution target detection technology.
□ Additionally, the industrial utility of the proposed technology is highly regarded because it does not need extra hardware resources and has a simple computational structure that supports real-time application.
□ Kim Sang-dong, lead researcher at the Division of Mobility Technology (also serving in the Interdisciplinary Engineering Major), stated, “This research improved the efficiency and precision of radar signal processing at the same time, enabling more accurate target detection without additional hardware. It is expected to develop into a core technology in national defense, autonomous driving, and unmanned systems.”
□ Meanwhile, this research, conducted by Kim Sang-dong and Kim Bong-seok’s team at DGIST and Professor Choi Yeong-doo’s team at the Republic of Korea Naval Academy, was funded through an institutional project at DGIST and the academic research program of the Naval Institute for Ocean Research at the Republic of Korea Naval Academy. The results were published in the November edition of the ‘Journal of Electrical Engineering & Technology,’ a globally recognized journal in the field of electrical and electronic engineering.
