The widespread use of low-cost, highly maneuverable unmanned aerial vehicles (UAVs), such as racing drones, has raised numerous security concerns. These UAVs commonly employ LoRa (Long Range) communication protocols, which feature long-range transmission and strong anti-interference capabilities. However, traditional countermeasure techniques targeting LoRa-based links often suffer from delayed response, poor adaptability, and high power consumption. To address these challenges, this study first leverages neural networks to achieve efficient detection and reverse extraction of key parameters from LoRa signals in complex electromagnetic environments. Subsequently, a continuous preamble injection jamming method is designed based on the extracted target signal parameters. By protocol-level injection, this method disrupts the synchronization and demodulation processes of UAV communication links, significantly enhancing jamming efficiency while reducing energy consumption. Experimental results demonstrate that, compared with conventional approaches, the proposed continuous preamble injection jamming method achieves improved signal detection accuracy, jamming energy efficiency, and effective range. To the best of our knowledge, this protocol-aware scheme, which integrates neural network-based signal perception and denoising, offers a promising and cost-effective technical pathway for UAV countermeasures.
Wu et al. (Fri,) studied this question.