AI-assisted frequency-modulated continuous wave radar for drone detection near runways: Challenges, trends, and research gaps

Abstract

Unmanned aerial vehicles (UAVs), birds, and foreign objects near airport runways pose a serious threat to aviation safety, particularly during takeoff and landing phases, when aircraft operate at high speeds and low altitudes. Although drone-related incidents remain infrequent, they can cause significant disruptions, including costly flight delays and safety concerns. This paper presents a focused review of drone detection techniques, emphasizing the integration of frequency-modulated continuous wave (FMCW) radar and artificial intelligence methods – fespe-cially convolutional neural networks – for UAV detection near runways. The article highlights the key benefits and limitations of these approaches in low-altitude, cluttered environments typical of airport infrastructure. Par-ticular attention is given to the role of digital signal processing, including the use of short-time Fourier transform for extracting micro-Doppler signatures from radar echoes, which are subsequently analyzed by deep learning models. The review also outlines current research gaps, including the lack of real-world data and scenario-specific studies. The aim is to support the development of reliable UAV detection systems tailored to the unique challenges of runway surveillance.